No new role cleared every hard gate on October 29. Three newly examined postings were rejected. GitLab’s Staff Solutions Architect role for AI is a pre-sales specialist position judged on technical wins and bookings. Parallel’s Forward Deployed Engineer role in Paris is mainly hospital IT deployment and integration. Idun Group’s Senior AI Engineer role for agentic AI pays from €55,000. Qonto’s Staff Software Engineer, AI Agents role (yesterday’s near-miss) still publishes no salary and is not repeated. Earlier picks and near-misses that are unchanged are not repeated either. These entries are research leads, not recommendations.
DecisionNo approval needed for this authorized daily website update. Zero new roles qualified today. Qonto’s Staff Software Engineer, AI Agents role still needs a published annual base salary (above €100,000, or at least €80,000 with equity) before reconsideration. No applications or outreach were made.
Zero roles passed every recorded gate on 2026-10-29. The entries below are research leads, not recommendations.
Uncertainty & caveats
Sources checked October 29, 2026. Posting dates come from employer ATS metadata or the employer page.
Unpublished pay is recorded as unknown, not as below the threshold. GitLab and Parallel publish no salary for these roles.
Qonto’s Staff Software Engineer, AI Agents (Python) (October 28 near-miss) is still listed with the same October 27 publish date and no compensation data. That decision stands, so it is not re-listed.
Inato’s Senior AI Engineer (October 26 pick) is still on its board with the same €85K–€100K plus equity range and October 23 publish date. It stays in the October 26 archive and is not counted today.
GitLab’s Forward Deployed Engineer – EMEA (October 19 near-miss) shows a refreshed October 28 update timestamp but still no published pay; a refreshed timestamp is not a material change. Shift Technology’s Lead AI Engineer and Staff AI Engineer still publish no salary.
Search leads not counted: a Dust LLM Engineer and a Mirakl Senior AI Developer appear with salary ranges only on a third-party job site and were not found on the employers’ own boards. Datadog’s Staff AI Engineer – Notebooks no longer appears on Datadog’s official board. Consulting roles at OCTO Technology, Equancy and a recruiter listing (Novera Talent) show ranges whose lower bounds are €54,000–€60,000 on third-party pages; they were not verified on employer pages and would fail the pay floor as quoted.
Agicap’s AI Software Engineer (Forward Deployment Engineer) is based in Lyon with a published €55,000–€70,000 range, created in 2023 per the ATS; it fails the location and pay gates and was not examined further.
Deduplication covered the accessible October 10–28 reports and the supplied October 3/5 Escape history.
Company financial diligence was not advanced because no role was ranked.
Independent recheck: every entry reflects checks made on 2026-10-29. Confirm the listing, pay and eligibility on the employer’s own pages before acting.
Automated official-ATS sweep on October 29: 132 of 276 attempted Ashby, Lever and Greenhouse board aliases returned data; many failed aliases are likely wrong board names, not evidence of no vacancies. The sweep screened AI-titled roles on reachable boards, Paris- or France-located engineering roles with published pay on reachable Ashby boards, and Greenhouse and Lever data for 108 France-located engineering roles (29 with pay text). It also screened Ashby engineering roles first published since October 26 with a France or Europe-remote location. Compared with October 28, the main sweep surfaced two newly matched senior AI-titled postings that list France, plus one reappearing Paris posting. A second pass reached 43 of 300 attempted aliases, and a Workable and Recruitee check reached 6 of 44. Five web searches were used for discovery only; leads found there were checked against employer pages and boards. Twelve listings or boards were examined individually on employer pages or official ATS data. Coverage is bounded, not exhaustive.
Criteria applied
Paris hybrid with explicit attendance, or fully remote explicitly allowing France-based applicants.
Senior/staff hands-on AI-central engineering; an AI employer, coding assistant use or AI product UI alone does not establish the required assignment.
Employer-published annual gross cash-base lower bound strictly above €100,000, or at least €80,000 with explicit equity. Use the lower bound; equity is not cash; packages, OTE, unlabelled pay ranges and estimates do not pass.
No unchanged prior selections. Requisition identity takes precedence over refreshed titles, URLs or timestamps; historical qualification is not current-live verification.
Filter entries
Showing 3 of 145 entries
JavaScript is off: all 145 entries from every report date are listed below.
Qualified matches
Entries whose stored gates all passed on their report date. Historical checks — not a live availability guarantee.
No role passed every recorded gate in the 2026-10-29 report. The research leads below are not recommendations.
No qualified entries in this view match the current filters.
Help Wanted
Ooak Data
Lead ML Engineer
Qualified
Report date
Evidence checked
Posted
Work location
Hybrid — “Location: Paris (city center), on-site with 1 to 2 days WFH”, which means three to four office days a week. The ATS summary separately labels the role “Occasional remote authorized”. No other country or remote option is offered.[1]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“Compensation: €90k to €120k + bonus + equity (BSPCE)”. The ATS summary gives “Salary: between 90000€ and 120000€ / year”, and the Y Combinator job page shows the same range. Bonus and equity are listed separately from the salary range. The listing does not say “gross” explicitly.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Paypass
Product researchpass
Gate notes
Pay passes the €80,000-plus-equity branch: the published annual salary floor is €90,000 and BSPCE equity is offered explicitly. Equity is not counted as cash, and the €90,000 floor does not clear the strict cash-only branch above €100,000.[1][2]
Location passes: the role is based at the Paris city-centre office, on-site with one to two work-from-home days a week.[1]
AI centrality passes: the lead owns machine learning across the data pipeline. That covers detecting sensitive entities in text and images, entity resolution, the knowledge graph that keeps pseudonyms consistent across tools, document understanding, and automated evaluation of anonymization and dataset quality. Next on the roadmap are reinforcement-learning environments and fine-tuning open-source models on company data.[1]
Seniority passes with a caveat: this is a hands-on lead role (“you prototype, code and ship to production yourself”), but it also includes people leadership. The lead grows the ML team from “a few ML engineers today” to about six by mid-2027 and runs 1:1s and hiring. The listing asks for 8+ years of ML engineering, including 2+ years leading ML engineers.[1]
Open on the check date: the ATS data dates the posting to November 4, 2026, and the application form rendered. The listing says the start is “as soon as you are available”, while the ATS summary shows a start date of January 4, 2027. The same role is on Ooak Data's Y Combinator job page.[1][2]
New to this site: no Ooak Data role appears in earlier reports.
Role/product noteOoak Data describes itself as an applied AI research lab that turns real company data into reinforcement-learning (RL) environments where AI agents learn to do real work. It connects to tools such as Slack, Gmail, Notion, Jira, SharePoint and Teams. An automated multimodal anonymization pipeline then turns that data into a “digital twin” that keeps the structure and relationships while transforming names, dates and proprietary content. From the twin, Ooak Data generates multi-step, multi-tool tasks calibrated against the latest frontier models. Its listed customers are frontier AI labs, enterprise AI teams and AI startups.[3] The careers page names the dataset effort Alexandria, described as a library of real-world business workflow datasets.[1] The role's AI work is the anonymization and data-understanding layer: sensitive-entity detection in text and images, cross-tool entity resolution, document understanding and automated quality evaluation, with RL environments and fine-tuning planned next.[1] Concrete engineering problems implied by the listing: keeping pseudonyms consistent across many tools without leaking real identities, measuring anonymization quality automatically so delivery can scale without manual checks, and handling partners' most sensitive data. Lab contracts and customers are company claims; no customer is named.[1][3]
Company researchInvestment: Ooak Data is in Y Combinator's Summer 2026 batch, founded in 2024.[4] The careers page says the company was “100% bootstrapped” before joining Y Combinator.[1] The amount of any investment is not publicly verified. Financial condition: the company's Launch YC post says it works with “multiple frontier labs” and is “now well above 7 figures in revenue”.[5] The job listing claims “8 figures contracts signed with frontier AI labs”.[2] Both are unaudited self-reports; profitability and runway are not publicly verified. Size: the Y Combinator profile lists a team size of 5, which may lag current hiring. The listing mentions “a few ML engineers today”.[1][4] Recent material activity: the Y Combinator batch, plus several open Paris roles on its Y Combinator profile, including ML, software and data engineering.[4]
Source evidence notes (5)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteLocation: Paris (city center), on-site with 1 to 2 days WFH. Compensation: €90k to €120k + bonus + equity (BSPCE). Stay hands-on: you prototype, code and ship to production yourself.
Evidence noteWe are a Y Combinator company with 8 figures contracts signed with frontier AI labs. Compensation: €90k to €120k + bonus + equity (BSPCE).
Evidence noteWe're working with multiple frontier labs and are now well above 7 figures in revenue.
Help Wanted
Inato
Senior AI Engineer
Qualified
Report date
Evidence checked
Posted
Work location
Remote — “PARIS OFFICE or REMOTE FROM FRANCE”; ATS location type: Remote. The application form adds: “This role requires you to be based in France and already have full authorization to work here.”[1][2]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“€85K – €100K • Offers Equity”. The official ATS data lists €85,000–€100,000 as a one-year EUR Salary component, with equity as a separate component. The listing does not say “gross” explicitly. This site’s archived October 14 report recorded the same requisition at €70K – €95K.[1][3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Paypass
Product researchpass
Gate notes
Pay passes the €80,000-plus-equity branch: the published annual salary floor is €85,000 and equity is explicitly offered as a separate component. Equity is not counted as cash, and the €100,000 ceiling does not clear the strict cash-only branch.[1][3]
Material change since this site’s history: the same requisition was rejected on October 14 because its published floor was €70,000. The employer republished it on October 23 with a €85,000–€100,000 range; the location wording and the duty quoted on October 14 are unchanged. The earlier range is in this site’s archived October 14 report.[1][3]
Location passes: the role is open to the Paris office or fully remote from France. The application requires living in France at least six months a year, paying tax there and already holding full French work authorization.[1][2]
AI centrality passes: the engineer orchestrates models over multimodal data, builds RAG over long medical documents and patient histories, and builds agentic systems for patient recruitment and site selection.[1]
Seniority passes on the title. The listing asks for several years with ML or AI at the centre of the work and shipped, maintained LLM features in production; it says AI work limited to prototypes, or a first AI feature shipped in the last year, is probably not a fit.[1]
Language requirement: the application says strong English and French are both assessed, and the engineering team works fully in French. The application form rendered on the check date.[2]
Role/product noteInato runs a platform that connects pharmaceutical trial sponsors with clinical research sites; its homepage claims 6,000+ sites and 25+ sponsors in 50+ countries (company claims). Its AI products cover site selection and pre-screening that automates chart review against trial eligibility criteria.[6] The listing names two example projects, which are examples rather than guaranteed assignments.[1] The first is a pre-screening tool built on reasoning models and visual RAG, used across hundreds of clinical sites. Inato’s March 2025 arXiv paper describes that pipeline: reasoning LLMs read unprocessed EHR documents as images, with multimodal embeddings for record search. It reports 93% criterion-level accuracy on the n2c2 2018 benchmark and 87% on a real-world set of 485 patients from 30 sites against 36 trials, with chart review taking under 9 minutes per patient on average.[4] The second is a site-selection agent that turns a sponsor’s request into constraints and objectives and calls a mathematical optimizer. An Inato engineer’s November 2025 post describes building it, including flaky tests from non-deterministic LLM calls, checking tool-call sequences instead of only final answers, and inspecting traces with Langfuse.[5] Concrete engineering problems: eligibility reasoning when records are incomplete (the paper attributes the real-world accuracy gap partly to this), evaluation of multi-call agents, and handling patient data, which Inato says is anonymized.[4][5]
Company researchInvestment: Inato announced a $20 million Series A2 on March 29–30, 2023, led by returning investor Cathay Innovation with Obvious Ventures, La Maison and Top Harvest Capital; MedCity News reported this brought total funding to $36 million.[8][9] No later funding round was found on the company’s news page or in this run’s searches, so newer financing is not publicly verified. Financial condition: revenue, profitability and runway are not publicly verified; funding does not establish them. Size: a current, attributable employee count is not publicly verified. Recent activity: Inato’s news page lists a September 2, 2026 Fierce AI Innovation Awards win in clinical trial operations.[7] pharmaphorum reported in February 2026 that Brad Cottreau joined as chief commercial officer.[10] The homepage features an October 23, 2026 recap of a SCOPE EU fireside chat with Sanofi.[6]
Source evidence notes (10)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence notePARIS OFFICE or REMOTE FROM FRANCE. Compensation €85K – €100K • Offers Equity. Build agentic systems to revolutionize key aspects of drug development such as patient recruitment or site selection
Evidence noteMetadata summary: Salary component €85,000–€100,000, interval 1 YEAR, currency EUR; separate equity component; published October 23, 2026; workplace type Remote; location PARIS OFFICE or REMOTE FROM FRANCE.
Evidence notean agent for clinical trial managers working at pharmaceutical companies to select the best set of clinical sites for a given clinical trial
Evidence noteBrad Cottreau was appointed chief commercial officer at Inato
Help Wanted
Langfuse (ClickHouse)
Langfuse - Senior Software Engineer, SDK
Qualified
Report date
Evidence checked
Posted
Work location
Remote — Location: Germany; France; Italy; Portugal; Spain; The Netherlands. ATS location type: Hybrid. Langfuse careers page: “European roles are remote-first with 3-5 days every two months in Berlin (travel covered); SF and Berlin locals use the office when it's useful.”[1][3]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“Salary €90K – €160K • Offers Equity”; the official ATS data classifies the €90,000–€160,000 component as an annual EUR salary. The listing adds that every new team member receives stock options. One published range covers all six listed countries.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Paypass
Product researchpass
Gate notes
Pay passes the €80,000-plus-equity branch: the published annual salary floor is €90,000 and equity is explicitly offered. Equity is not counted as cash, and the €90,000 floor alone does not clear the €100,000 cash-only branch.[1]
Location passes: France is one of six named countries, and Langfuse states that European roles are remote-first. The role still requires 3–5 days in Berlin every two months, with travel covered, and the ATS labels it “Hybrid”.[1][3]
AI centrality passes: the role owns Python and TypeScript SDKs that instrument LLM applications and agent loops, maintains integrations with 40+ AI frameworks and model providers, and supports evaluation workflows such as the experiment runner. The craft is SDK and API engineering rather than model training.[1][4]
Seniority is explicit in the title. It was first published on October 19, 2026, and the same-page application form rendered on the check date.[1][2]
Role/product noteLangfuse is an open-source (MIT) LLM engineering platform for tracing, evaluation and prompt management, used by teams building AI applications and agents. Its SDKs are built on OpenTelemetry, so third-party instrumentations appear inside Langfuse traces, and they also expose the prompt-management and evaluation features.[4] This role owns those SDKs: reliability and low CPU and memory overhead inside customers’ production systems, from serverless functions to long-running agent loops; batching, async flushing and graceful degradation; framework integrations; the OTLP endpoint; and contributing to OpenTelemetry GenAI semantic conventions.[1] A concrete near-term problem is migration: Langfuse’s docs say the legacy POST /api/public/ingestion endpoint on Langfuse Cloud stops accepting everything except scores on November 16, 2026.[4] A narrower assigned project beyond SDK ownership is unknown.
Company researchLangfuse was acquired by ClickHouse, announced on January 16, 2026, alongside ClickHouse’s $400 million Series D led by Dragoneer; the acquisition price was not disclosed.[5] TechCrunch, citing Bloomberg, reported a $15 billion valuation for that round.[7] ClickHouse reported more than 3,000 ClickHouse Cloud customers and ARR growth above 250% year over year; these are company claims, not audited financial statements.[5] Langfuse said it had term sheets for a Series A before choosing the acquisition and that it remains open source and self-hostable.[6] Langfuse ended 2025 with over 20K GitHub stars and 26M+ SDK installs per month, per ClickHouse.[5] Revenue, profitability, runway and Langfuse team headcount are not publicly verified.
Source evidence notes (7)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteSalary €90K – €160K • Offers Equity. We maintain SDKs in Python and TypeScript both built on OpenTelemetry. Langfuse integrates with 40+ frameworks and model providers. Equity in the company - Every new team member who joins our company receives stock options.
Evidence noteEuropean roles are remote-first with 3-5 days every two months in Berlin (travel covered); SF and Berlin locals use the office when it's useful.
Evidence noteThe Langfuse SDKs are built on top of OpenTelemetry. On Langfuse Cloud, POST /api/public/ingestion stops accepting everything except scores on November 16, 2026.
Evidence noteannounced today the close of its Series D financing, raising $400 million. The company now serves more than 3,000 customers on its fully managed service, ClickHouse Cloud, with ARR growing more than 250 percent year over year.
Evidence noteDatabase provider ClickHouse secured $400 million at a $15 billion valuation, Bloomberg reported
Help Wanted
OpenAI Deployment Company
Forward Deployed Engineer - EU
Qualified
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris, France. Location Type: Hybrid. “This role is based in the France, with hybrid work and customer travel shaped by deployment needs.” No weekly office-day count is published.[1]
Salary evidence
Known annual EUR base
Equity
Yes
Pay noteL3-L5: €115K – €200K • Offers Equity. Official ATS data lists €115,000–200,000 as a one-year Salary component, separate from equity; the employer says ranges may change during the interview process.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Paypass
Product researchpass
Gate notes
Pay passes: the published annual salary floor of €115,000 is strictly above €100,000, and equity is explicit. Equity was not counted as cash.[1]
Location passes on employer evidence: Paris, France with an explicit Hybrid location type. The office-day cadence and customer-travel share are not stated, so check them before applying.[1]
Seniority and AI centrality pass: the employer seeks a Senior Applied AI Software Engineer to build agentic LLM applications and own evaluation and production readiness.[1]
The application form rendered on the check date. It asks about work authorization in the job country and states a limit of three applications per 180 days.[2]
Role/product noteProduct: DeployCo is OpenAI’s enterprise deployment arm. Forward Deployed Engineers work directly with customers to put AI systems into production, then turn repeatable patterns into product capabilities such as the Agent SDK and model benchmarking and reliability tools.[4] Users and problem: large enterprises that need to move from AI experiments to reliable production workflows. One customer example on the official site is BBVA, which is scaling to 120,000 employees across 25 countries.[4] Duties: lead engineering workstreams for customer deployments, owning architecture, implementation quality, evaluation strategy and production readiness for LLM-powered systems. Prototypes become reusable reference architectures.[1] AI mechanism: agents, retrieval, orchestration, evaluation and monitoring built on closed and open-source foundation models, including OpenAI models and APIs.[1] Named project: unknown, because the work depends on the customer. Interestingness: Medium-High. The concrete engineering problems are production agent reliability, evaluation design and generalising customer work into reusable components. The caveat is that this is consulting-style delivery with customer travel.
Company researchInvestment: announced May 11, 2026 with more than $4 billion of initial investment from a 19-partner group led by TPG, with Advent, Bain Capital and Brookfield as co-leads.[3][5] Ownership: majority-owned and controlled by OpenAI.[3] Size: the announced Tomoro acquisition was to bring about 150 Forward Deployed Engineers and Deployment Specialists from day one. The official announcement made it subject to closing conditions; completion and current headcount were not verified from an official source in this run.[3] Disclosed financial condition, revenue, profitability and runway: not publicly verified. Funding commitments do not establish profitability.
Source evidence notes (5)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is based in the France, with hybrid work and customer travel shaped by deployment needs. We are seeking a Senior Applied AI Software Engineer to lead the design, build, and deployment of enterprise AI systems. L3-L5 €115K – €200K • Offers Equity.
Evidence noteIt will launch with more than $4 billion of initial investment. DeployCo is majority-owned and controlled by OpenAI. The acquisition will bring approximately 150 experienced Forward Deployed Engineers and Deployment Specialists to DeployCo from day one.
Role/product notePlanning and performance software for finance, HR, sales and supply-chain teams. This role improves deployed multi-agent features over a deterministic, auditable computation engine. A narrower assigned project is unknown.
Company researchEmployer-reported 600+ staff and nearly $400m raised. March 2026 company reporting said ARR was approaching $100m; September coverage reported a later company claim that ARR exceeded $100m. These are not audited financial statements. Profitability and runway are not publicly verified.
Source evidence notes (5)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteFrance; Remote; €80,000–€140,000 annual salary and stock options meet the €80k-plus-equity rule.; Multi-agent feature engineering is central; a separate product source was read.
Role/product noteAI-assisted accounting ERP for mid-market finance teams. The role owns agent workflows, RAG, experimentation and production infrastructure. A separate MCP product article documents permissions and draft → validate → approve → save. It does not establish that this hire would build the MCP server.
Company researchCompany announcement of 2 October 2025 reported $90m Series A and over $100m total funding, thousands of users and $100bn in journal entries processed. These are dated company claims, not verified revenue or profitability. August 2026 MCP launch is recent product activity. Current size, revenue, newer funding and runway are not publicly verified.
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteRemote (EU, LATAM); Remote-first team; Base salary $140,000–$200,000 and separately stated equity $45,000–$65,000.; At the 9 October 2026 ECB rate EUR1=USD1.1206, the base floor is about EUR124,933; this is not a guaranteed euro-denominated French offer.; Agent workflows, RAG and production evaluation are central; separate product evidence read.
Evidence noteHistorical value recorded by 10 October research: 9 October 2026 EUR1=USD1.1206. This endpoint changes daily; current content is not the historical snapshot.
Research leads
Rejected and near-miss entries kept for research. These are not recommendations.
No research leads recorded in the 2026-10-29 report.
No research entries in this view match the current filters.
Wanted
Scaleway
Software Engineer AI Agents
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris”, with Toulouse, Rouen, Lyon, Bordeaux, Lille and Rennes also listed; Lever workplace type “hybrid”. The benefits section says “Hybrid work: Up to 3 days remote per week”, so at least two office days a week are implied.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing, and the Lever job data has no salary range. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity term.[1][2]
AI centrality passes: the team is building a new AI Agentic Platform so customers can build, deploy and operate AI agents at scale; the engineer designs and develops the platform API, runs it in production and works with the AI Inference team on Scaleway's LLM catalog.[1]
Seniority is not in the title, but the requirements open with “Senior profile with extensive experience in Software Engineering, complex systems architecture”. The listed stack is Go, Kubernetes applications and operators, REST/gRPC APIs, PostgreSQL and agent frameworks such as LangChain or CrewAI.[1]
New requisition: the Lever posting was created on November 17, 2026, and no Scaleway role has appeared in earlier reports. The listing and its application form loaded on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Scaleway is a European cloud provider (bare metal, containers, serverless and AI products), backed by the Iliad Group.[1] Its Generative APIs product serves AI models through OpenAI-compatible, pay-per-token APIs, with guides for retrieval-augmented generation and tool-calling agents.[3] This role builds the next layer: a managed platform for building, deploying and operating AI agents and multi-agent workflows on top of that model catalog.[1]
Company researchListing-stated only: the Iliad Group is “investing €3 billion” in a sovereign AI alternative.[1] Scaleway's own revenue, headcount and funding were not publicly verified in this run.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteServe the latest AI models via API, pay by million token
Wanted
Anthropic
[Paris] Applied AI Architect, Industries
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris, France”. “Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time.”[4]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay noteThe listing shows “Annual Salary: €190.000 — €220.000 EUR” but adds that “For sales roles, the range provided is the role’s On Target Earnings ("OTE") range”. This pre-sales role has no separate base figure, so the range is not confirmed as cash base.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Hands-on AI-engineering gate fails: the employer describes “a pre-sales architect focused on becoming a trusted technical advisor”, working with account executives from technical discovery to deployment; the build work is the customer's.[4]
Pay gate unverified: the published range may be on-target earnings including sales commission, not base salary.[4][5]
The listing asks for 5+ years in technical customer-facing roles such as Solutions Architect or Sales Engineer, and native or fluent French (C1 or above). Paris and minimum 25% office attendance are explicit; the same-page application form loaded on the check date.[4]
New requisition, first published November 18, 2026. Anthropic's similar Paris Applied AI Architect, Partnerships role was rejected on October 16 for the same pre-sales reason.[4][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Anthropic builds the Claude models, sold through the Claude API and Claude for Work. This role helps large enterprises evaluate and integrate Claude. Separate product diligence was not done because the role failed the hands-on engineering gate.[4]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: first published November 18, 2026; location Paris, France; pay range €190,000–€220,000 with the sales OTE note.
Wanted
Mistral AI
Applied AI Engineer, CyberSecurity, Defense
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris”, employment type “Full time”. No location type (on-site, hybrid or remote) and no office-attendance rule is published; the Ashby job data has an empty workplace type.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing, and the Ashby job data has no compensation tier. The benefits section names healthcare, parental leave and allowances, not pay. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity term.[1][2]
Location gate unverified: Paris is the only location named, but no work mode or attendance rule is stated.[1][2]
Eligibility is narrower than usual: the listing limits applicants to French nationals (“France for now”) who can meet local requirements for high-level security clearance, and asks for fluent French.[1]
AI centrality passes: the engineer composes and deploys agentic cyberdefense workflows (alert triage and investigation, incident-response automation, detection engineering, threat-intelligence and log analysis), designs evaluations for correctness and safety, and owns deployment including on-premises and air-gapped setups. Pre-sales calls are part of the role. Seniority is not in the title; the listing asks for 5+ years building cybersecurity systems plus 1+ year of hands-on agentic or LLM work.[1]
New requisition, first published November 17, 2026. It is separate from Mistral AI's Applied AI, Forward Deployed Engineer – CyberSecurity (Europe) requisition listed as a near-miss on November 4, which is still on the board. The listing and its application tab loaded on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral AI makes frontier and open-weight models, Forge for customising them, and Studio for building and governing agents, with deployment on-premises, in a customer's VPC or hosted by Mistral.[3] This role builds AI-for-cyberdefense systems on that stack for defence, government and critical-infrastructure customers, often in sovereign, on-premises or classified environments, and feeds field patterns back into Mistral's products.[1]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence notecomposing cybersecurity-related agentic workflows focused on cyberdefense, configuring harnesses for real-world scenarios, and delivering value directly to clients
Evidence noteRun on-prem, in your VPC, or have us run it for you.
Wanted
Kraken Technologies
Senior Software Engineer - AI Foundations
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “London, UK; Berlin, Germany; Manchester, UK; Paris, France”, location type “Hybrid”. No office-day count is published. The listing also states that visa sponsorship is not offered and applicants must already have the right to work in the country they apply to.[4][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing, and the Ashby job data has no compensation tier. Pay is unknown, not below the threshold.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity term.[4][5]
Location gate unverified: Paris is one of four hybrid locations, but no attendance rule is published.[4]
Seniority and AI centrality pass: a hands-on senior individual-contributor role in the team that lets other teams at Kraken build and run AI agents. The engineer builds Python services such as gateways, agent runtimes, knowledge pipelines and eval harnesses, and turns permissions, guardrails, budgets and data-access rules into software. The listing also asks for AWS and Kubernetes operations experience.[4]
First published August 25, 2026 (not recent) and new to this site; it surfaced today through a job-aggregator search, and the employer listing and its application tab loaded on the check date.[4][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Kraken sells an operating system for energy utilities, with clients including EDF, Octopus Energy and National Grid; its home page reports 90m+ customer accounts in 15+ countries and promotes AI for utility customer service.[6] The AI Foundations team runs internal AI infrastructure: a knowledge layer that grounds LLMs in company context, an AI hub with an MCP gateway, a LiteLLM gateway behind coding agents, and evals, guardrails and observability for live AI runs.[4]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked. The home page claims 90m+ customer accounts (employer statement).[6]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: published August 25, 2026; location London with Manchester, Paris and Berlin as secondary locations; workplace type Hybrid; no compensation tier.
Evidence noteBoosting customer service with Utility-Grade AI
Wanted
Photoroom
Product ML Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris”, with “Hybrid working from our Paris headquarters”. No office-day count is published.[7][8]
Salary evidence
No published figures
Equity
Yes
Pay noteThe listing offers “Competitive salary + meaningful equity” but publishes no salary figure, and the Ashby job data has no compensation tier. Equity is stated; the cash base is unknown, not below the threshold.[7][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Seniority gate fails: the listing asks for 2+ years of hands-on experience shipping ML-powered products and has no senior or staff level.[7]
Pay gate unverified: equity is stated but no base salary is published.[7][8]
AI centrality passes: the engineer owns short ML projects inside a product team, choosing between fine-tuning vision or diffusion models and integrating external models, then ships, evaluates and monitors them in production.[7]
New requisition ID, first published November 15, 2026. Photoroom requisitions with the same title were rejected on October 13 and October 27 for the same seniority reason; the listing and its application tab loaded on the check date.[7][8]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Photoroom sells AI photo and video editing for e-commerce through a mobile app, web platform and API, and says it processes around 7 billion images a year. Separate product diligence was not done because the role failed the seniority gate.[7]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: published November 15, 2026; location Paris; no workplace type; no compensation tier.
Wanted
Filigran
Principal Software Engineer - XTM One
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “France; Germany; Italy; Portugal; Spain; United Kingdom”, location type “Remote”, and “Remote-first flexibility — work from wherever you’re based”. France is named explicitly.[1][2]
Salary evidence
No published figures
Equity
Yes
Pay noteThe listing offers “Competitive compensation + equity — everyone shares in our success” but publishes no salary figure, and the Ashby job data has no compensation tier. Equity is stated; the cash base is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: equity is stated, but no employer-published base salary exists for either branch of the rule.[1][2]
Location passes: a remote requisition with France named among six countries.[1][2]
Seniority passes (Principal, 10+ years in software engineering with technical leadership, an individual-contributor role with no headcount). AI centrality passes: the engineer owns the architecture of XTM One's agent runtime and orchestration, tool and MCP integrations, LLM latency and inference cost, and the evaluations and guardrails for agent behaviour in production, and writes production code.[1]
New requisition, first published November 12, 2026 and not in earlier reports on this site; the listing and its same-page application tab loaded on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Filigran builds an open-source threat-management suite (OpenCTI for threat intelligence, OpenAEV for attack simulation and exposure validation, OpenGRC). XTM One is its agentic AI layer: pre-built and custom security agents (threat hunting, attack-scenario creation, remediation guidance) that work across those products with human approval steps, bring-your-own-LLM support, and SaaS or on-premises deployment. This role sets XTM One's technical direction across all connected products.[1][3]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked. The listing states 60+ engineers at Filigran (employer statement).[1]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteOwn the platform architecture: agent runtime and orchestration, tool and MCP integrations, data and context layers, APIs, permission model, shared commons.
Evidence noteMetadata summary: published November 12, 2026; location France with Germany, Portugal, Spain, United Kingdom and Italy as secondary locations; workplace type Remote; no compensation tier.
Evidence noteXTM One is the agentic AI layer that consolidates and orchestrates all assistive and agentic AI capabilities across the entire XTM platform and supercharges threat-informed exposure management – so you can move from threat signal to validated action at machine speed.
Wanted
Datadog
Staff Research Scientist - Post-Training for Observability Native Models
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris, France”, tagged “#LI-Hybrid”, and “We operate as a hybrid workplace to ensure our Datadogs can create a work-life harmony that best fits them.” No office-day count is published.[4][6]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing, and the official Greenhouse job data has an empty pay-range field. Pay is unknown, not below the threshold.[4][6]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity term.[4][6]
Location gate unverified: Paris-based and described as hybrid, but no attendance rule is published.[4]
Seniority and AI centrality pass: a Staff, hands-on individual-contributor research role that designs post-training methods, simulated environments, synthetic data and evaluation frameworks for decision models and agents operating in observability and security settings. The listing asks for a PhD or equivalent experience and a research-impact track record.[4]
New requisition, first published November 13, 2026; the listing and its embedded Greenhouse application form loaded on the check date.[4][6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Datadog sells an observability and security platform. Datadog AI Research (DAIR) builds models on telemetry data; its public work includes Toto 2.0, a family of open-weights time-series forecasting models (4 million to 2.5 billion parameters) trained on observability and synthetic data.[7] This role post-trains decision and generative models for agents that act across complex systems in observability and security.[4]
Company researchNot publicly verified in this run; Datadog is a listed company, but no filing, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteBuild simulated environments, synthetic data generation approaches, and evaluation frameworks that enable scalable model training and rigorous measurement
Evidence noteMetadata summary: requisitions 8385200 and 8385205 were first published November 13, 2026, located “Paris, France”, with an empty pay-range field. The Greenhouse application form is embedded on each careers page.
Evidence noteToday we’re releasing Toto 2, a family of open-weights time series forecasting models, on Hugging Face.
Wanted
Datadog
Staff Research Scientist - Foundation & World Models
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris, France”, tagged “#LI-Hybrid”, and “We operate as a hybrid workplace to ensure our Datadogs can create a work-life harmony that best fits them.” No office-day count is published.[5][6]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing, and the official Greenhouse job data has an empty pay-range field. Pay is unknown, not below the threshold.[5][6]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity term.[5][6]
Location gate unverified: Paris-based and described as hybrid, but no attendance rule is published.[5]
Seniority and AI centrality pass: a Staff, hands-on individual-contributor research role that trains large multimodal foundation models on metrics, logs, traces, topology and events, covering pre-training, representation learning, scaling and evaluation. The listing asks for a PhD or equivalent experience and a research-impact track record.[5]
New requisition, first published November 13, 2026; the listing and its embedded Greenhouse application form loaded on the check date.[5][6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Datadog sells an observability and security platform. Datadog AI Research (DAIR) builds models on telemetry data; its public work includes Toto 2.0, a family of open-weights time-series forecasting models (4 million to 2.5 billion parameters) trained on observability and synthetic data.[7] This role extends that foundation-model work to multimodal telemetry and world models of distributed systems.[5]
Company researchNot publicly verified in this run; Datadog is a listed company, but no filing, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteTrain large-scale multimodal models on diverse telemetry data, including metrics, logs, traces, topology, events, and other non-text modalities
Evidence noteMetadata summary: requisitions 8385200 and 8385205 were first published November 13, 2026, located “Paris, France”, with an empty pay-range field. The Greenhouse application form is embedded on each careers page.
Evidence noteToday we’re releasing Toto 2, a family of open-weights time series forecasting models, on Hugging Face.
Wanted
ElevenLabs
Application Security Engineer - Agents platform
Rejected
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “United Kingdom; Dublin; London”, location type “Remote”, and “This role is remote and can be executed globally. However, to facilitate working with the Security Team, we prefer candidates based in GMT to GMT+3 or UK.” France is not named.[8][9]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing or in the Ashby job data. Pay is unknown, not below the threshold.[8][9]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Location gate fails as ambiguous: the listed locations are the UK, Dublin and London, and “can be executed globally” does not explicitly name France as an eligible country.[8]
Pay gate unverified: no employer-published base salary or equity term.[8][9]
AI centrality is partial: the engineer builds security tooling, authentication and authorization controls and guardrails for ElevenAgents, a platform for conversational AI agents; security for AI and LLM applications is listed as a bonus, not a requirement. The listing asks for 5+ years in application security engineering.[8]
New requisition, first published November 13, 2026; the listing loaded on the check date.[8][9]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: ElevenLabs builds AI voice models and three platforms, including ElevenAgents for deploying voice and chat agents. Separate product diligence was not done because the role failed the location gate.[8]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is remote and can be executed globally. However, to facilitate working with the Security Team, we prefer candidates based in GMT to GMT+3 or UK.
Evidence noteMetadata summary: published November 13, 2026; location United Kingdom with Dublin and London as secondary locations; workplace type Remote; no compensation tier.
Wanted
GitLab
Staff Forward Deployed Engineer, EMEA
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: “Remote Ireland; Remote, Austria; Remote, France; Remote, Germany; Remote, Netherlands; Remote, Spain; Remote, United Kingdom”, and “All of our roles are remote, however some roles may carry specific location-based eligibility requirements.” France is named explicitly.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing, and Greenhouse's pay-transparency data for this job is empty. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[1][2]
Location passes on employer wording: remote, with France named among the seven eligible countries.[1]
Seniority passes (Staff title; the listing asks for the “technical standards and contribution discipline of a Staff-level product engineer”). AI centrality is partial: driving production adoption of GitLab Duo Agent Platform is one of four duty areas, alongside production code contributions, strategic customer troubleshooting and thought leadership. The listing asks for strong Ruby on Rails or Go.[1]
New requisition, first published November 12, 2026; the same-page application form loaded on the check date. A similar GitLab role (Staff Forward Deployed Engineer, Agentic SDLC, requisition 8522265002) was a near-miss here on October 19; that requisition's job data now returns “not found”.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: GitLab sells a DevSecOps platform. GitLab Duo Agent Platform is its agent layer: built-in and custom agents and flows for planning, coding, security analysis and pipelines, triggered from GitLab events, with policy controls over which agents and models may run and a trace of every agent action. This role helps strategic customers adopt it safely and feeds what it learns back into the product. Specific customer projects are not named.[1][3]
Company researchNot publicly verified in this run; GitLab is a listed company, but no filing, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteAs a Staff Forward Deployed Engineer, you'll work closely with customers while maintaining the technical standards and contribution discipline of a Staff-level product engineer. Drive governed, measurable, and production-oriented adoption of GitLab Duo Agent Platform with strategic customers.
Evidence noteMetadata summary: first published and updated November 12, 2026; location lists seven remote countries including France; the pay-range field is empty.
Evidence noteGitLab Duo Agent Platform is the repo-side AI that carries the change through review, tests, security, and pipelines, to a change that actually ships.
Wanted
Mistral AI
Research Engineer - Tokenizer & Protocols
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: “This role is primarily based at one of our European offices (Paris and London).” Remote is conditional: “In certain specific situations, we will also consider remote candidates based in one of the countries listed in this job posting — currently France & UK”, with the first week and then “at least 4 days per month” in the local office. No office-day rule is stated for office-based hires.[4][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing or in the Ashby job data. Pay is unknown, not below the threshold.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[4][5]
Location gate unverified: the role is office-based in Paris or London; remote work from France is allowed only “in certain specific situations”, and no weekly attendance rule is given for office-based hires.[4]
Seniority gate unverified: the title and listing state no level or years of experience.[4]
AI centrality passes: the engineer designs, builds and maintains the tokenization and model-interface libraries used across pretraining, post-training and deployment, so that training, evaluation and production use the same formats.[4]
New requisition, first published November 12, 2026; the listing and its application tab loaded on the check date.[4][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral AI builds frontier and specialist open-weight models that customers can use as-is or adapt, plus Studio for building, governing and improving agents and applications, and Vibe for research, coding and everyday tasks.[6] This role owns the tokenizers, chat formats and model-interface libraries that, in the listing's words, sit “between a conversation and a model”; the listing says these decisions reach every model Mistral ships.[4]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign, build and maintain core libraries for tokenization and model interfaces. In certain specific situations, we will also consider remote candidates based in one of the countries listed in this job posting — currently France & UK.
Evidence noteMetadata summary: both Mistral AI requisitions below were first published November 12, 2026, with Paris as the primary location and no compensation tier. Tokenizer & Protocols has no workplace type; Agentic Engineering is marked Remote, with Linz, Munich, London and Berlin as secondary locations.
Evidence noteStart with Mistral’s frontier and specialist open-weight models. Use them as they are, or make them specific to your organization. Build, govern and improve agents and applications with Studio.
Wanted
Mistral AI
AI Scientist - Agentic Engineering
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing header: “Paris; Berlin; Linz; London; Munich”, location type “Remote”. The listing text states no remote-eligibility or office-attendance policy.[7][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing or in the Ashby job data. Pay is unknown, not below the threshold.[7][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[7][5]
Location gate unverified: the job data marks the role Remote and lists Paris among five cities, but the listing gives no written remote-eligibility or attendance rule.[7][5]
Seniority gate unverified: no level or years of experience are stated.[7]
AI centrality passes: the role designs pretraining, SFT and RL data and verifiers for engineering tasks (CAD, CAE, EDA), builds evaluation benchmarks, and designs agent architectures and harnesses for long multi-step workflows. It asks for demonstrated experience with real CAD, CAE, semiconductor-simulation or EDA workflows.[7]
New requisition, first published November 12, 2026, in the same AI4Engineering area as the “AI Scientist, Agentic CAD” requisition listed here on November 3 (which Mistral re-published the same day). The listing and its application tab loaded on the check date.[7][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral AI builds frontier and specialist open-weight models that customers can use as-is or adapt, plus Studio for building, governing and improving agents and applications, and Vibe for research, coding and everyday tasks.[6] This role is in AI4Engineering Science, which trains Mistral's models to reason about and carry out engineering tasks in CAE and EDA/semiconductor tools.[7]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign and improve agent architectures and harnesses: how models plan, call tools, recover from errors, and chain steps together across long-horizon engineering workflows
Evidence noteMetadata summary: both Mistral AI requisitions below were first published November 12, 2026, with Paris as the primary location and no compensation tier. Tokenizer & Protocols has no workplace type; Agentic Engineering is marked Remote, with Linz, Munich, London and Berlin as secondary locations.
Evidence noteStart with Mistral’s frontier and specialist open-weight models. Use them as they are, or make them specific to your organization. Build, govern and improve agents and applications with Studio.
Wanted
Neural Concept
Forward Deployed Engineer - France
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris”, location type “Hybrid”, and “Balance life and work with a hybrid model and flexible hours”. No office-day rule is stated.[8][9]
Salary evidence
No published figures
Equity
Unknown
Pay noteThe listing offers “a competitive salary” but publishes no figure or equity term, and the Ashby job data has no compensation tier. Pay is unknown, not below the threshold.[8][9]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[8][9]
Location gate unverified: Paris hybrid, but no attendance rule is published.[8]
Seniority passes: 7+ years in engineering, simulation or applied machine learning is a “strict requirement”. AI centrality passes: the engineer applies the company's machine- and deep-learning workflows to customers' engineering and physics problems and builds proofs of concept on its platform. French and English are both mandatory.[8]
First published November 9, 2026 and not in earlier reports on this site; the listing and its application tab loaded on the check date.[8][9]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Neural Concept sells an AI platform for product engineering teams that uses 3D deep learning on CAD and simulation data to speed up design in fields such as aerodynamics, thermal management and structural mechanics, working with CAD tools like CATIA and NX and simulation tools like Fluent and Abaqus. Its forward deployed engineers run customer projects in the automotive, aerospace and energy industries.[8][10]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked. The homepage claims use by “50% of the world’s leading OEMs and top-tier suppliers”; that claim was not independently checked.[10]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteYou have 7+ years of professional experience in engineering, simulation, or applied machine learning, ideally in an industrial or customer-facing environment (strict requirement)
Evidence noteNeural Concept is trusted by 50% of the world’s leading OEMs and top-tier suppliers to drive performance breakthroughs and stay ahead of the curve.
Wanted
Seedtag
Senior Agentic Platform Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: “Multiple locations · Fully Remote”, with a “Flexible work model with 100% remote or hybrid options. (Remote contracts available in Spain, Italy, UK, Belgium, Netherlands, France, and Germany.)” France is named explicitly; the application form also lists Paris among its location choices.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary or equity term is published on the listing. The application form asks “What are your yearly salary expectations?” for the “expected annual fixed salary (gross)”. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published salary or equity, so neither the €100,000 branch nor the €80,000-plus-equity branch can be checked.[1][2]
Location passes on employer wording: fully remote, with remote contracts explicitly available in France.[1]
AI centrality and seniority pass: the engineer owns key parts of Liz Agent, Seedtag's AI agent platform: the agent runtime and context management, MCP tool servers, multi-model routing, memory, evaluation and prompt-injection protection. The listing asks for 5+ years building production backend systems and hands-on experience shipping LLM applications or AI agents.[1]
Open on the check date: the listing and its Teamtailor application form loaded. The page's structured data gives a posting date of October 16, 2026; no Seedtag role appears in earlier reports on this site.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Seedtag sells contextual advertising for the open web and connected TV that does not rely on cookies or user tracking. Its proprietary AI is called Liz. Liz Agent, launched on March 11, 2026, is a conversational agent for brands and agencies that plans and activates media campaigns from Seedtag's data. Seedtag says it runs on “a sophisticated multi-agent orchestration engine” combining LLMs with Seedtag's own data and tools. The listing says the platform runs on 30+ services on Google Kubernetes Engine, with Python and Go, MCP, PydanticAI, LangGraph, the Claude Agent SDK and OpenAI Agents SDK, Langfuse and LiteLLM.[1][3]
Company researchEmployer-stated on the listing: founded in 2014, 500+ employees in 17 countries and “backed by €250M in funding”. These figures were not independently verified. Revenue and profitability were not publicly verified.[1]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteRemote contracts available in Spain, Italy, UK, Belgium, Netherlands, France, and Germany. As our Senior Agentic Platform Engineer, you will take technical ownership of key components of Liz Agent, Seedtag's AI agent platform. 5+ years of experience building and maintaining production backend systems.
Evidence noteSummary: the application form loaded on the check date. It asks “What are your yearly salary expectations?” and offers Paris among its location choices.
Evidence noteLiz Agent is built on a sophisticated multi-agent orchestration engine that harmonizes state-of-the-art Large Language Models (LLMs) with Seedtag’s exclusive data, tools, and deep domain expertise.
Wanted
Alice & Bob
Senior MLOps Engineer (MLOps / LLMOps)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: “Paris”, “CDI / Hybrid”, with a “Flexible remote policy, up to 40 % a month”, so most working time is in the office. The application form asks whether the applicant holds a valid work permit for France.[4][5]
Salary evidence
No published figures
Equity
Yes
Pay noteNo salary is published. The benefits list says “own it with our BSPCE plan” (French startup stock options) and mentions bonuses for patents. The application form asks for the “expected salary range for this role (annual gross in EUR)”. Pay is unknown, not below the threshold.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: equity (BSPCE) is stated, but no cash-base figure is published, so the €80,000-plus-equity branch cannot be checked.[4][5]
Location passes on employer wording: Paris hybrid with remote work capped at 40% a month.[4]
AI centrality and seniority pass, with a caveat: the engineer builds and runs the company's MLOps and LLMOps infrastructure on Google Cloud (pipelines, deployment, monitoring) for both classic ML and GenAI/LLM use cases. The role supports the company's own researchers; it is not work on a customer-facing AI product. The listing asks for 7+ years in ML engineering, MLOps or software engineering.[4]
Open on the check date: the listing and its Lever application form loaded. The Lever posting data shows it was created in August 2026. It has not appeared in earlier reports on this site; it appeared in today's sweep because the board answered today.[4][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Alice & Bob is building a fault-tolerant quantum computer based on “cat qubits”, a superconducting qubit design that it says corrects some errors by itself. This role belongs to its Data & AI Ops team, which is building an “AI-ready Data Platform” on Google Cloud. The work turns research ML and AI experiments into reproducible, monitored production systems, and helps teams across the company adopt AI tooling.[4][6]
Company researchThe company homepage says NVIDIA's NVentures has joined its Series B; the round size and date were not checked. The listing describes a team of 250+ people. Revenue and profitability were not publicly verified.[4][6]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteAs a Senior MLOps / LLMOps Engineer, you will help turn ML and AI experiments into reliable, production-ready solutions. Design reliable and scalable solutions for both traditional ML and GenAI / LLM use cases. Our success is your success: own it with our BSPCE plan. Flexible remote policy, up to 40 % a month.
Evidence noteSummary: the application form loaded on the check date. It asks “What is your expected salary range for this role (annual gross in EUR)?” and whether the applicant holds a valid work permit for France.
Evidence noteNVIDIA’s NVentures joins our Series B. We pioneer the cat qubit, the first qubit with built-in error correction.
Wanted
Cast AI (Kimchi)
Senior Applied ML Engineer | Kimchi (Model Router)
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “Austria; France; Germany; Italy; Netherlands; Poland; Spain; United Kingdom”, with a “flexible, remote-first global environment” and a #LI-Remote tag. France is named explicitly. The listing also says Cast AI provides no visa sponsorship or work permit.[1]
Salary evidence
No published figures
Equity
Yes
Pay noteThe listing offers “Competitive salary (depending on the level of experience)” and “Equity options”, but publishes no figure, and the official Greenhouse job data has no pay range. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: equity options are stated, but no cash-base figure is published, so neither the €100,000 branch nor the €80,000-plus-equity branch can be checked.[1]
Location passes on employer wording: a remote role whose location list names France.[1]
AI centrality and seniority pass: the listing calls it “an applied-ML seat, not an infrastructure one”. The engineer owns the model inside Kimchi's LLM router (training data, objective, evaluation, calibration and release) and asks for 5+ years in applied ML or ML research.[1]
Open on the check date: the listing loaded with an embedded Greenhouse application form. First published October 15, 2026; no Cast AI or Kimchi role appears in earlier reports on this site.[1]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Kimchi is Cast AI's autonomous multi-model coding agent and inference service. An orchestration engine routes each task to the best-fit model on complexity and cost, mostly to self-hosted open-weight models, and keeps frontier models for the hardest steps. This role owns the routing model itself: for each request it predicts the quality, cost and decoding speed each candidate model would deliver. The listing names production-parity evaluation with error bars, shadow and A/B rollouts, drift monitoring and calibrated abstention as part of the work. Cast AI's own claim of 2.5x lower cost than a commercial-models-only baseline in shadow-mode evaluations was not independently checked.[1][2]
Company researchNot publicly verified in this run. No funding, revenue or headcount source was checked because the role did not pass the pay gate.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis is an applied-ML seat, not an infrastructure one. Competitive salary (depending on the level of experience). Equity options. Enjoy a flexible, remote-first global environment. Location: Austria; France; Germany; Italy; Netherlands; Poland; Spain; United Kingdom.
Evidence noteKimchi is an autonomous multi-model coding agent. The orchestration engine routes every task to the best-fit model on complexity and cost.
Wanted
Axelera AI
Senior/Staff AI Application Engineer - Heterogeneous Computing
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location type “Remote”; the location list includes “France (remote)” and “Paris (on-site)”, and the role can be done by working “fully remotely from any European country (incl. the UK) you are already in”. The listing adds that “priority will be given to candidates who are [interested in being] based in Belgium or Italy”.[3]
Salary evidence
No published figures
Equity
Yes
Pay noteThe listing describes “an attractive compensation package, including a pension plan, extensive employee insurances and the option to get company shares”, but publishes no figure. The application form asks for salary expectations. A €120,000–€160,000 range appears only on a job aggregator and was not counted. Pay is unknown, not below the threshold.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published salary; aggregator figures are not employer evidence.[3][4]
Location passes on employer wording: France is listed as a remote location. Applicants should note the stated priority for Belgium- or Italy-based candidates.[3]
AI centrality and seniority pass: the engineer builds the model-enablement platform that decides how each PyTorch or ONNX model is split across Axelera's AI accelerator, the host CPU and other accelerators, plus an agent harness for model bring-up. The listing asks for 8+ years in systems, compilers, runtimes or performance.[3]
Open on the check date: the listing and its Ashby application form loaded. First published October 26, 2026; no Axelera role appears in earlier reports on this site.[3][4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Axelera AI designs AI accelerator chips and cards (Metis today, Europa named in the listing) for vision, agentic and generative AI workloads, from embedded modules to servers, using digital in-memory computing. The listing ties the role to the end-to-end performance customers see in Axelera's Voyager SDK. This role builds the platform and agent harness that partition, compile, run and profile models across the accelerator and host, including large models, cascades and multi-model pipelines. It is model-systems and compiler work rather than LLM application building.[3][5]
Company researchNot publicly verified in this run. No funding, revenue or headcount source was checked because the role did not pass the pay gate.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteFrance (remote) is among the listed locations. Work fully remotely from any European country (incl. the UK) you are already in. We offer an attractive compensation package, including a pension plan, extensive employee insurances and the option to get company shares.
Evidence noteSummary: the Ashby application form loaded on the check date and asks for “Salary expectations (total yearly gross compensation, in your local currency)”.
Evidence noteVision, Agentic and Gen AI running on the world's most efficient silicon, from embedded to edge to server.
Wanted
Qonto
Senior AI Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris, France; Barcelona, Spain; Berlin, Germany; Madrid, Spain; Milan, Italy”, location type “Hybrid”. No office-day count is published; the application form asks which working arrangement the applicant wants and says it “will be discussed during the hiring process”.[6][7]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary is published on the listing or in the official Ashby job data. Qonto's careers page says recruiters discuss compensation early in the process and that ESOPs are offered “for eligible roles”, which does not confirm equity for this role. Pay is unknown, not below the threshold.[6][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published salary or role-specific equity.[6][8]
Location gate unverified: Paris-based and marked hybrid, but no attendance rule is published.[6][7]
AI centrality passes: the engineer builds customer-facing agents and designs their evaluation, observability and tooling. Seniority is uncertain: despite the senior title, the requirements read as mid-level and accept products shipped “during an internship or as a side project”.[6]
New requisition first published November 10, 2026; the listing and its application form loaded on the check date. Earlier Qonto AI requisitions on this site (October 12–28) are different roles.[6][7]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Qonto offers business banking and finance tools to small businesses. This role sits in its AI Products team, which ships agents that “take real actions for customers”, and focuses on making them reliable: sizing evaluation to the risk of each action, building production monitoring, and turning evaluation and monitoring into a self-service platform for other teams. The listed stack is Python with Pydantic AI, models accessed through AWS, Grafana for usage and cost per sub-agent, and a self-hosted Braintrust instance for evaluation.[6][8]
Company researchNot publicly verified in this run. No funding, revenue or headcount source was checked because the role did not pass the pay and location gates.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteYou'll design the evals, observability and tooling that keep agent quality high in a regulated bank. Build agents: You'll develop agents that take real actions for customers. Location type: Hybrid.
Evidence noteSummary: the application form loaded on the check date and asks “What working arrangement are you ideally looking for? (we will discuss it during the hiring process)”.
Evidence noteYour recruiter will discuss compensation early in the interview process. ESOPs (Employee Stock Ownership Plans) are offered for eligible roles.
Wanted
Datadog
Senior AI Research Engineer - LLM Training & Inference Infrastructure
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris, France”, tagged “#LI-Hybrid”, and the benefits mention collaborating with “colleagues across the Datadog offices in New York City and Paris”. No office-day count or remote eligibility is published.[1][2]
Salary evidence
No published figures
Equity
Yes
Pay noteNo salary range is published on the listing, and the official Greenhouse job data has no pay range for this requisition. The benefits list “New hire stock equity (RSUs) and employee stock purchase plan (ESPP)”. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published base salary. Equity (RSUs) is stated, but the €80,000-plus-equity branch still needs a published cash floor.[1][2]
Location gate unverified: Paris-based and tagged hybrid, but no attendance rule is published.[1]
AI centrality and seniority pass: a senior research-engineering role that designs, builds and operates distributed training and inference infrastructure for large language models, including Mixture-of-Experts models, and builds “execution environments and sandboxes for online RL and agent post-training”.[1]
New requisition first published November 9, 2026; the listing and its Greenhouse application page were reachable on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Datadog describes itself as an observability and security platform for the AI era. Both roles sit in Datadog AI Research (DAIR), on the Frontier Foundations team, which builds the infrastructure to train and serve large language models. That team supports DAIR's work on trained agents for observability: post-training models to operate autonomously across Datadog's domain, with SRE incident response named as the first target and code repair, security response and infrastructure optimization as later ones. The listings name earlier AI work (Bits AI, Bits Evolve and the Toto time-series foundation model). Separate company or product diligence beyond the listings was not done because neither role passed the hard gates.[1][3]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked because no role was ranked.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteBuild execution environments and sandboxes for online RL and agent post-training. New hire stock equity (RSUs) and employee stock purchase plan (ESPP).
Evidence noteSummary: Paris, France; first published November 9, 2026; no pay range in the official job data; application form present.
Wanted
Datadog
Senior AI Research Engineer - GPU Systems & Kernels
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris, France”, tagged “#LI-Hybrid”. No office-day count or remote eligibility is published.[3][4]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing, and the official Greenhouse job data has no pay range for this requisition. Pay is unknown, not below the threshold.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published base salary or equity.[3][4]
Location gate unverified: Paris-based and tagged hybrid, but no attendance rule is published.[3]
AI centrality and seniority pass: the engineer will “write, tune, and debug GPU kernels (CUDA, Triton, or similar) for the operations that dominate LLM training and inference”, reduce communication cost in distributed training, and profile training and rollout workloads.[3]
New requisition first published November 9, 2026; the listing and its Greenhouse application page were reachable on the check date.[3][4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Datadog describes itself as an observability and security platform for the AI era. Both roles sit in Datadog AI Research (DAIR), on the Frontier Foundations team, which builds the infrastructure to train and serve large language models. That team supports DAIR's work on trained agents for observability: post-training models to operate autonomously across Datadog's domain, with SRE incident response named as the first target and code repair, security response and infrastructure optimization as later ones. The listings name earlier AI work (Bits AI, Bits Evolve and the Toto time-series foundation model). Separate company or product diligence beyond the listings was not done because neither role passed the hard gates.[3]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked because no role was ranked.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteSummary: Paris, France; first published November 9, 2026; no pay range in the official job data; application form present.
Wanted
Neuralk
Machine Learning Engineer (Senior, Data Science Agent)
Near-miss
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — Official listing: “You will be located in our Paris offices” (page location “Paris / London”). Neuralk's careers page describes a hybrid model “with up to two days of remote work per week”, which means at least three office days.[5][6]
Salary evidence
No published figures
Equity
Yes
Pay noteThe employer listing offers “A competitive salary” and “Equity (BSPCE)” but publishes no figure. A €110,000–€140,000 range appears only on a job aggregator and was not counted. Pay is unknown, not below the threshold.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: the employer page states equity but no salary figure; aggregator figures are not employer evidence.[5]
Location passes on employer wording: Paris office with up to two remote days a week.[5][6]
AI centrality and seniority pass: the senior engineer will “help design and build the machine learning intelligence behind our Data Science Agent”, including the agent's decision logic for exploring, training, comparing and evaluating models on tabular data. The listing asks for 7+ years of applied ML or ML engineering.[5]
Open on the check date: the listing shows “Open” and its application form loads. The employer page shows no posting date; aggregators first listed it on November 5, 2026. No Neuralk role appears in earlier reports on this site.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Neuralk builds foundation models for structured (tabular) business data and a Data Science agent powered by them, which helps data scientists from problem framing to production-ready models on problems such as churn prediction, risk modelling and demand forecasting. This role builds the agent's modelling logic, predictive pipelines and evaluation protocols on top of the research team's models; the listing says it is not core model research or infrastructure operations. Separate company or product diligence beyond the employer pages was not done because the role did not pass the pay gate.[5][6]
Company researchThe listing calls Neuralk an “early-stage, well-funded AI startup”; that is a company claim. Funding amounts, revenue and headcount were not publicly verified in this run.[5]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence notehelp design and build the machine learning intelligence behind our Data Science Agent. A competitive salary Equity (BSPCE). You will be located in our Paris offices.
Evidence noteWe offer a flexible hybrid work model that lets you enjoy the vibrant energy of our office as often as you like, with up to two days of remote work per week.
Evidence noteSummary: the listing's “Apply here” button opens this Machine Learning Engineer application form, which loaded on the check date.
Wanted
Vibe
Staff Full Stack Engineer - AdManager
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — ATS location: Paris, hybrid. The listing says: “We're in the Heart of Paris and our team is in 3x a week!”[6]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“The pay range for this role is €99,000 - €133,000.” The ATS lists the range as a one-year EUR salary component, with equity (an employee stock ownership plan) and a bonus as separate components.[6]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Paypass
Product researchunknown
Gate notes
AI centrality fails: the role builds the advertiser-facing campaign setup, reporting, creatives, integrations, payments and account management. AI appears only as the team's AI-assisted development workflows, testing design-to-code tools, and “Built AI-powered products” as a nice-to-have.[6]
Pay and location would pass: a €99,000 floor with explicit equity, and Paris hybrid with three office days a week.[6]
New listing first published November 8, 2026; the listing and application tab were reachable on the check date.[6]
Evidence noteThe pay range for this role is €99,000 - €133,000. Hybrid flexibility: our team is in 3x a week. The team uses AI-assisted development workflows.
Wanted
Belmont Lavan
Agent Security Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official Workable listing: “Hybrid · Paris, Île-de-France, France”, full time. The page's structured job data also marks the job as telecommute with applicants required to be in France. No office-day count is stated.[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published; the listing's salary fields are empty. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published base salary or equity.[1]
Location gate unverified: Paris hybrid limited to France-based applicants, but no attendance rule is published, and the page's “hybrid” label conflicts with its telecommute job data.[1]
AI centrality passes: the role secures autonomous AI agents before and after production, designs guardrails, attacks the systems itself and builds governance controls. It asks for 5+ years in security engineering, including at least 1 year securing LLM or agent-based systems, and hands-on knowledge of prompt injection, jailbreaks, tool-based data exfiltration and RAG poisoning. No seniority level is in the title.[1]
Product gate unverified: the listing does not say whether the hire secures Belmont Lavan's own agents or client deployments, and no agent product or project is named.[1][2]
New listing first posted November 4, 2026; the listing and its Workable application form were reachable on the check date.[1]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Belmont Lavan describes itself as a system integrator for startups and corporates, designing and implementing ERP, CRM, EAM, open-source, Microsoft and cybersecurity systems, with AI and ML among its listed capabilities. The listing describes agent identity, least-privilege access, monitoring and testing, but names no client, product or agent platform.[1][2]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked because no role was ranked.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteBelmont Lavan is a trusted leader in IT solutions, specializing in the design, development, and implementation of robust ERP, CRM, EAM, Open Source, Microsoft, and Cybersecurity systems.
Unknown — The employer page now shows only “Paris | Headquarters” and no job description, so no current attendance rule could be read.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary is shown on the employer page. A €120,000–€180,000 figure seen on a job aggregator was not counted because the employer did not publish it.[3]
Recorded gates (as checked on the report date)
Listing openfail
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Open gate fails: the employer page says “This position is no longer active. Either the position was filled, or the ad has expired.” The role is not in Trustpair's current list of open positions.[3][4]
Found through a job aggregator that still showed the role as posted November 6, 2026, with pay the employer never published. The employer page's structured data dates the posting to November 27, 2025.[3]
AI centrality was doubtful in any case: discovery copies describe a Ruby on Rails distributed-systems role in which LLM experience is only a bonus.[3]
Evidence noteSummary: the open-positions list on the check date includes Senior Backend Engineer (Ruby on Rails) and Engineering Manager roles but no Staff Software Engineer.
Wanted
Mistral AI
Forward Software Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris”, full time, Solutions department. No location type, office-day rule or remote option is stated in the listing or the job data.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the official job data. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[1][2]
Location gate unverified: Paris-based, but no hybrid attendance rule or remote eligibility is published.[1][2]
AI centrality and seniority pass: a hands-on Applied AI role that designs agent architecture (orchestration, tool use, memory, retrieval), builds security into agents (permissions, sandboxing, prompt-injection defenses, audit trails) and owns task-level evals and tracing. It asks for “5+ years building production software, with at least one agentic or GenAI system shipped to production that you ran after launch.”[1]
New requisition first published November 6, 2026; it is separate from Mistral AI’s Applied AI Forward Deployed Engineer - CyberSecurity role (near-miss on November 4), which is still listed. The listing and its application form were reachable on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral AI builds frontier and open-weight models plus tools to customize them, Studio for building and governing agents and applications, and hosted or on-premises inference. This role sits in Applied AI, works with customer-facing teams and turns what works for one customer into reusable patterns; the listing says security is “a big part of where we're heading” and names alert triage, threat hunting and investigation agents as relevant experience. Specific customer projects are not named.[1][3]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today because no role was ranked.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: both Forward Software Engineer requisitions published November 6, 2026, location Paris, no compensation tier; the Data variant is marked Hybrid with London as a secondary location, the other has no workplace type.
Evidence noteBuild, govern and improve agents and applications with Studio.
Wanted
Mistral AI
Forward Software Engineer, Data
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris; London”, location type “Hybrid”. The number of office days is not stated.[4][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the official job data. Pay is unknown, not below the threshold.[4][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[4][2]
Location gate unverified: hybrid in Paris or London, but no attendance rule is published.[4]
AI centrality and seniority pass: builds AI systems end to end with a data-engineering focus, including distributed pipelines for retrieval, agentic workflows and fine-tuning, and an evaluation loop with LLM and human labelling. It asks for 5+ years building and running production data-intensive pipelines.[4]
New requisition first published November 6, 2026; the listing and its application form were reachable on the check date.[4][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: the same Applied AI team as the Forward Software Engineer role, focused on the data architecture under customer AI systems: ingestion, ontologies and data models, governance (ownership, access control, lineage, retention) and versioned evaluation datasets. Mistral AI’s product line covers models, model customization, agent tooling and inference hosting.[4][3]
Company researchNot publicly verified in this run.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteBuild distributed pipelines that process large volumes of documents, code, logs, and multimodal content for retrieval, agentic workflows, and fine-tuning.
Evidence noteMetadata summary: both Forward Software Engineer requisitions published November 6, 2026, location Paris, no compensation tier; the Data variant is marked Hybrid with London as a secondary location, the other has no workplace type.
Evidence noteBuild, govern and improve agents and applications with Studio.
Wanted
Abbeal (recruiter for an unnamed legal-tech client)
Senior LLM Engineer — Agents IA en production, legal-tech
Near-miss
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — Recruiter listing: “Hybride Paris : 2 à 3 jours par semaine au bureau, dans Paris centre”, with one to two weeks on site at the start.[5]
Salary evidence
Known annual EUR base
Equity
Yes
Pay noteRecruiter listing: “CDI, 90 K€ de fixe plus des BSPCE dès l'arrivée.” (permanent contract, €90K fixed salary plus founder share warrants from the start). These terms are stated by the recruiter, not by an employer page.[5]
Recorded gates (as checked on the report date)
Listing openunknown
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Employer verification gate unmet: the page belongs to Abbeal, an engineering and technical-recruitment firm, and the client is not named, so no employer listing, application path or company source could be checked.[5]
Pay and location would pass if the employer confirmed them: €90,000 fixed plus BSPCE, and 2–3 office days per week in central Paris (recruiter statements).[5]
AI centrality and seniority pass on the recruiter’s description: a senior individual-contributor role designing and running a production agent graph with many tool calls, fallbacks, trace analysis and versioned evaluation datasets. Posting date not published; the page and its application form were reachable on the check date.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteRecruiter description only: an unnamed Paris legal-tech company building an AI agent for notaries and real-estate professionals (sale deeds, estates, powers of attorney, email) that the recruiter says is in paid production with thousands of monthly users and six engineers. Not independently verified.[5]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteCDI, 90 K€ de fixe plus des BSPCE dès l'arrivée.
Wanted
Beyond Skills (recruiter for an unnamed AI-infrastructure startup)
Senior AI Engineer
Near-miss
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — Recruiter listing: “Sur site à Paris, avec un jour de télétravail par semaine” (on site in Paris with one remote day per week); only Paris-based applicants are considered.[6]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay noteRecruiter listing: “Rémunération : 80 à 120 k€+ selon profil, plus de l'equity”. The range is not labelled as fixed base salary, and the terms are stated by the recruiter, not by an employer page.[6]
Recorded gates (as checked on the report date)
Listing openunknown
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Employer verification gate unmet: Beyond Skills is a recruitment firm and the client is not named, so no employer listing, application path or company source could be checked.[6]
Pay gate unverified: €80,000–€120,000+ plus equity is stated as “rémunération” (pay), not explicitly as base salary. The four-office-day rule is also a recruiter statement.[6]
AI centrality and seniority pass on the recruiter’s description: the hire owns model benchmarking and evaluation, prompt caching and the core inference-cost optimisation algorithm, and needs 5+ years including production AI agents. Posting date not published; the page and its application form were reachable on the check date.[6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteRecruiter description only: an unnamed, recently seed-funded Paris startup whose product helps software publishers get more from AI models while cutting inference costs; product in beta with first customers. Not independently verified.[6]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteRémunération : 80 à 120 k€+ selon profil, plus de l'equity
Wanted
EverAI
Mid/Senior AI LLM Engineer (Full Remote - France)
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “France; Portugal”, location type Remote, “Work From Anywhere: Fully remote.” The company says it prefers B2B contracts but is flexible.[7][8]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay note“Annual Compensation Range (Gross) €75K – €125K.” The listing says “Published ranges reflect total gross annual compensation, including any applicable target bonus.” No equity is stated.[7]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Pay gate fails: the range is total compensation including target bonus, and its €75,000 lower bound is below both branches of the rule.[7]
Location passes: a fully remote requisition explicitly named for France.[7][8]
AI centrality passes: fine-tuning self-hosted open-weight LLMs, building training data and evals, and shipping model improvements to production. This is a new France-specific requisition first published November 6, 2026, with the same pay as the Europe-wide version rejected on October 24.[7][8]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: EverAI builds AI companion products and says most of its traffic runs on self-hosted, fine-tuned open-weight models; it reports 50M+ users (employer statement, not independently verified). The listing says the work involves explicit content.[7]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location France with Portugal as a secondary location, workplace type Remote, published November 6, 2026, salary component €75,000–€125,000 per year.
Wanted
Believe
Staff Engineer Agentic AI
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Official listing: location “Paris, France”; the SmartRecruiters job data marks the role neither remote nor hybrid, while the benefits section mentions a “Remote working policy”. No number of office or remote days is stated, so the attendance rule is unknown.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the official job data. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[1][2]
Location gate unverified: Paris-based, but no explicit hybrid attendance rule is published.[1][2]
AI centrality and seniority pass: a staff-level, hands-on role building Believe’s enterprise agentic framework (orchestration, retrieval, tool and MCP integration, human-in-the-loop checkpoints) and its eval, guardrail and tracing layer. The listing asks for “5+ years in software engineering, including a meaningful stretch at a cross-team level of influence (Staff, Principal or equivalent), and recent experience taking LLM-based systems into production”.[1]
First released September 8, 2026 and not previously examined in this archive; the listing page and its “I’m interested” application button were reachable on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Believe is a music company that “supports the world’s music makers and rights holders at every stage of their journey”, working with artists, songwriters, labels and publishers. This role sits in Core Engineering beside the Agentic & Automation team and builds shared agentic services used by engineering, product, data and business teams across the group; the stated stack is Python, Java, GCP (including Vertex AI), Kubernetes, Kafka, MySQL, MongoDB, Snowflake, Spark, KrakenD and n8n. The specific internal agents to be built are not named.[1][3]
Company researchNot publicly verified in this run. The listing describes Believe as a simplified joint-stock company under French law; no funding, revenue, profitability or headcount source was checked today.[1]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteAs Staff AI Engineer in Core Engineering, you help design and build the systems behind intelligent automation, agentic workflows and large-scale retrieval at Believe.
Evidence noteMetadata summary: name Staff Engineer Agentic AI, location Paris, France, remote false, hybrid false, released September 8, 2026, no compensation field.
Evidence noteBelieve supports the world’s music makers and rights holders at every stage of their journey.
Wanted
Oyster
Senior AI Product Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official job data: location “EMEA”, workplace type Remote, with secondary locations Romania, Hungary, Bulgaria, Austria, Ireland, Portugal, Latvia, Chile, Slovakia, Spain, Brazil, Poland, Colombia, Peru, Argentina, Germany and Greece. France is not listed.[4][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the official job data.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location gate fails: the fully remote role lists eligible countries, and France is not among them.[5]
Pay gate unverified: no published salary or equity.[4][5]
AI centrality and seniority pass: a senior, hands-on role owning Oyster’s in-product AI assistant end to end (agent tools, retrieval, evaluation loops, production root-cause fixes). First published November 5, 2026.[4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Oyster runs a global employment platform for hiring, paying and supporting staff in other countries; “Oyster AI” is its assistant for customers and team members. The backend is Ruby on Rails and the frontend React.[4]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteIn this role you’ll own complex improvements to it end to end, from defining the problem and designing the solution through evaluation, rollout and learning from production.
Evidence noteMetadata summary: location EMEA, workplace type Remote, published November 5, 2026, seventeen secondary locations that do not include France, no compensation tier.
Wanted
Deepgram
Senior Pre-Sales Solutions Engineer - Europe
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: “Note: this is a remote role that is to be based out of one of the following countries: United Kingdom, Ireland, Scotland, Switzerland”. France is not listed.[6]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the official job data.[6][7]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Location gate fails: remote work is limited to the United Kingdom, Ireland, Scotland and Switzerland.[6]
AI centrality gate fails: a pre-sales role whose duties are technical discovery, proofs of concept, demos and security reviews for sales opportunities, not building or operating AI systems as the primary duty.[6]
Pay gate unverified: no published salary or equity. First published November 5, 2026.[6][7]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Deepgram sells real-time speech-to-text, text-to-speech and voice-agent APIs, available as cloud APIs or self-hosted software.[6]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location EU - Remote, workplace type Remote, published November 5, 2026, no compensation tier.
Wanted
Databricks
AI Forward Deployed Engineer - France
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: “We welcome remote applicants located near our offices. Preferred locations: Paris, France. Remote location for candidates to be based in the France region only (not remote from any other country)”. The listing also asks for willingness to travel to customers once every 4–8 weeks.[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing, and the official Greenhouse data for this requisition carries no pay range. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published base salary or equity.[1][2]
Seniority is not established: the listing asks for “Min 4 years' experience” and names no level; the title carries no senior or staff designation.[1]
Location passes on employer evidence: remote work is explicitly allowed for candidates based in France, with Paris as the preferred location.[1]
AI centrality passes: duties include “Develop cutting-edge GenAI solutions” and “Own production rollouts of consumer and internally facing GenAI applications”, with RAG, multi-agent systems, Text2SQL and fine-tuning listed as expected experience. It is a customer-facing professional-services seat. First published November 4, 2026; the embedded Greenhouse application form rendered on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Databricks sells a data and AI platform; its Agent Bricks product is described as a “unified platform to build and govern AI agents”, with access to OpenAI, Anthropic, Google and open-source models, governance of agents, MCP servers, models and tools, and automatically generated synthetic data for agent evaluation. The AI Forward Deployed Engineering team delivers professional-services engagements that help customers build and put first-of-their-kind GenAI applications into production, and feeds field learning back to product and engineering.[1][3]
Company researchNot publicly verified in this run. The listing states that more than 20,000 organizations use the Databricks platform; no funding, revenue, profitability or headcount source was checked today.[1]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteUnified platform to build and govern AI agents that continuously improve.
Wanted
Mistral AI
Applied Scientist Lead (Robotics), EMEA
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris; Amsterdam; Lausanne; London”, location type “Hybrid”. No number of office days is stated.[4][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published, and the official Ashby data has no compensation tier. Pay is unknown, not below the threshold.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no employer-published salary or equity.[4][5]
Location is not established: the role is hybrid across four European cities, including Paris, but no office-attendance rule is stated.[4][5]
Seniority passes: “Mistral AI is seeking a Lead / Staff Applied Scientist to help lead our robotics research and industrial collaborations”. The role mixes hands-on model work with building and mentoring a team.[4]
AI centrality passes: the role designs, trains and deploys vision-language and vision-language-action models for robot navigation and manipulation, runs pre-training and post-training on GPU clusters, and builds evaluators and benchmarks. The listing asks for a track record of shipping models on real robots. First published November 4, 2026; the application form rendered on the check date.[4][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral builds frontier and specialist open-weight models and sells model customization (Forge), agent and application tooling (Studio, Vibe) and hosted or self-hosted inference and compute to enterprises and public bodies. This Applied Science seat sets technical direction for robotics and delivers robotics models with industrial partners, from scoping to deployment.[4][6]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today.
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMistral AI is seeking a Lead / Staff Applied Scientist to help lead our robotics research and industrial collaborations in the Applied Science team.
Evidence noteMetadata summary: location Paris with Amsterdam, London and Lausanne as secondary locations, workplace type Hybrid, published November 4, 2026, no compensation tier.
Evidence noteStart with Mistral’s frontier and specialist open-weight models. Use them as they are, or make them specific to your organization.
Wanted
Anthropic
Manager, Forward Deployed Engineering
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France. “Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.” The role also requires 25–50% travel to customer sites.[7]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay noteAnnual Salary: €235.000 — €295.000 EUR. The listing adds: “For sales roles, the range provided is the role’s On Target Earnings ("OTE") range”. Whether this figure is base or OTE is not stated for this role, and no equity grant is stated.[7]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Hands-on engineering gate fails: “This is a player-coach role with a strong bias toward leadership … your primary impact will come from the people you hire, the standards you set, and the culture you create.” The listing asks for 2+ years of people management.[7]
Pay is published but not confirmed as cash base, because the listing applies an OTE caveat to sales roles without saying whether it covers this one.[7]
Location passes: Paris with a stated minimum of 25% office time. First published November 4, 2026; the same-page application form rendered on the check date.[7]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: first manager of the Southern Europe Forward Deployed Engineering team, which embeds with strategic customers to ship production applications built on Anthropic’s Claude models.[7]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis is a player-coach role with a strong bias toward leadership.
Wanted
Moxie
Staff AI Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “Remote - Global”, location type “Remote”. Compensation is published only for Canada, Chile, Brazil and Colombia, and the Dominican Republic; the listing does not say that France-based applicants are eligible.[8][9]
Salary evidence
Known annual base, non-EUR (no conversion applied)
Equity
Unknown
Pay noteAnnual salary tiers: Canada $195K – $225K, Chile $137K – $170K, Brazil and Colombia $130K – $165K, Dominican Republic $97K – $113K (USD). No tier covers France and no equity is stated, so no France-applicable floor exists to convert to EUR.[8][9]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location gate fails: “Remote - Global” is not an explicit statement that France-based applicants are eligible, and the published pay tiers name only countries in the Americas.[8][9]
Pay gate unverified for France: no France tier and no equity are published.[8][9]
AI centrality and seniority pass: a staff-level hands-on IC role leading production LLM features, including an assistant that answers medical-spa texts and calls, and building evaluation systems. First published November 3, 2026; the application form rendered on the check date.[8]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Moxie provides software and services to independent medical-aesthetics practices; this role leads its production AI features (the Maia assistant, a context engine and AI-drafted messages) and its shared AI platform.[8]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location Remote - Global, workplace type Remote, published November 3, 2026, salary tiers for Canada, Chile, Brazil and Colombia, and the Dominican Republic only.
Wanted
Yield Studio & Advisory
Lead AI Engineering
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Employer listing on Welcome to the Jungle: “CDI · Paris · Télétravail fréquent” (permanent contract, Paris, frequent remote work). The application page shows “Paris · Hybrid”. The number of office days is not stated.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteThe listing states “Salaire : non spécifié” (salary not specified), and no equity term is published. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchpass
Gate notes
Pay gate unverified: no salary figure or equity is published.[1]
Location is not established: the role is in Paris with “frequent remote work”, but no office-day rule is stated.[1][2]
AI centrality passes: “Très hands-on, vous construisez notre Software Factory agentique : une plateforme où des agents IA prennent en charge une part croissante du cycle de développement” (hands-on, you build our agentic software factory, a platform where AI agents take on a growing share of the development cycle), including multi-agent architectures, context engineering and agent orchestration.[1]
Seniority passes: the listing asks for a “Profil Senior, Staff ou Principal Engineer”. Part of the role is internal enablement (training teams, setting AI-assisted development standards), not only building. First published November 2, 2026; the application form rendered on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Yield Studio is a product and software agency that designs and builds web, mobile, business software and AI systems for client companies, and says an internal Lab builds AI tools for its own delivery. This role builds that internal agentic “software factory”, where agents cover specification, design, development, testing, code review, deployment and maintenance, and changes how the company’s product, design and engineering teams work with AI.[1][3]
Company researchNot publicly verified. The employer’s Welcome to the Jungle profile shows 30 employees and a 2019 founding year; no funding, revenue or profitability figure was checked.[1]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteTrès hands-on, vous construisez notre Software Factory agentique : une plateforme où des agents IA prennent en charge une part croissante du cycle de développement (spécification, conception, développement, tests, code review, déploiement, maintenance).
Evidence noteNotre Lab’ interne développe les outils IA qui éliminent le travail répétitif : scaffolding intelligent, génération de boilerplate, pipelines auto-configurés.
Unknown — Official listing: location “Paris”, employment type “Full time”. The official Ashby data records no workplace type, and the listing states no remote or office-attendance rule.[4][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published, and the official Ashby data has no compensation tier. Pay is unknown, not below the threshold.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary or equity.[4][5]
Location is not established: the listing names Paris but no workplace type or attendance rule.[4][5]
Seniority is not established: the listing asks for “2+ years as a technical individual contributor (data scientist or software engineer) on AI-based products” and names no level.[4]
AI centrality passes: the role composes “cybersecurity-related agentic workflows” and configures harnesses for agentic vulnerability scanning, red-teaming and penetration-testing scenarios for clients. It also includes pre-sales calls and may cover non-security use cases. First published July 9, 2026 and not previously examined on this site; the application form rendered on the check date.[4][5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral provides frontier models, developer tools, applications and compute to enterprises in finance, manufacturing, defense, healthcare and the public sector. This forward-deployed seat builds client use cases on the Mistral stack, with cybersecurity assignments as the priority: cloud-defense harness configuration, agentic vulnerability scanning, agentic red-teaming and penetration-testing scenarios.[4]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is about composing cybersecurity-related agentic workflows, configuring harnesses for real-world scenarios, and delivering value directly to clients.
Evidence noteMetadata summary: location Paris, no workplace type, published July 9, 2026, no compensation tier.
Wanted
vector8
Senior AI/ML Engineer, France
Near-miss
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — Official listing header: “Permanent · Hybrid · France, France”; body: “The role is primarily based in Paris with a proportion of the time spent working with stakeholders on client sites.” The number of office or client-site days is not stated.[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteThe employer listing says only “Competitive salary package”, with no figure and no equity term. A third-party job site shows €100,000–€150,000, but that figure is not employer-published and was not used; a €100,000 floor would also not be strictly above €100,000 without explicit equity. Pay is unknown.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: the employer publishes no salary figure or equity.[1]
Location is not established: the listing is hybrid and “primarily based in Paris”, but it states no office-day rule and includes client-site time.[1]
AI centrality passes: “Design and build agents and AI workflows, including tool integration, orchestration, guardrails and human validation steps” and “Integrate LLMs, retrieval and classical ML or speech models with clients' existing systems and data.”[1]
Seniority passes (Senior; 5+ years in AI/ML engineering, software development or a related field). The listing requires fluent English and French. New to this site’s history; the employer page shows no posting date. The application tab on the same listing rendered a form on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: vector8 is an AI consultancy that designs, develops and delivers production AI solutions for clients, from agentic systems to LLM, classical ML and speech solutions, with a focus on regulated industries (security, compliance, access control, legacy integration). This role builds client agents and workflows, evaluation, tracing and MLOps.[1]
Company researchNot publicly verified in this run. The careers page lists open roles in France, Switzerland, the UK and North Macedonia; no headcount, funding or revenue figure was checked.[2]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteSummary: the open-roles list links to the Senior AI/ML Engineer, France requisition on HiBob.
Wanted
Mistral AI
AI Scientist, Agentic CAD (AI4Engineering)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris; Amsterdam; Berlin; Linz; London; Munich; Palo Alto”, location type “Hybrid”. No office-day rule is stated.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published, and the official Ashby data has no compensation tier. Pay is unknown, not below the threshold.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary or equity.[3][4]
Location is not established: Paris is one of seven listed hubs and the role is hybrid, but no attendance rule is stated.[3]
AI centrality passes: the role is about “teaching Mistral's LLMs to generate, analyze, and modify CAD designs”, building pre- and post-training data, verifiers and agent harnesses.[3]
This is a research-scientist seat that asks for “Demonstrated hands-on CAD experience” as well as agentic-system and reward-modeling experience, so it is narrower than a general AI engineering role. First published November 2, 2026 per the official Ashby data. The application form rendered on the check date.[3][4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mistral builds frontier models, developer tools, applications and compute. The AI4Engineering Science team builds data and evaluation infrastructure and runs SFT/RL for models that do engineering work; this seat covers text-to-CAD and image-to-CAD generation, CAD part analysis and design modification, plus agent scaffolding for multi-step CAD workflows.[3]
Company researchNot publicly verified in this run; no funding, revenue or headcount source was checked today.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteYour focus is a single mission: teaching Mistral's LLMs to generate, analyze, and modify CAD designs the way an expert design engineer would.
Evidence noteMetadata summary: location Paris, workplace type Hybrid, published November 2, 2026, no compensation tier.
Wanted
Veepee
Senior AI Engineer - H/F/X
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing header: “Paris”, “VeepeeTech – Engineering / Permanent / Hybrid”; benefits: “a hybrid setup with up to 2 days of remote work per week”, so at least three office days a week.[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range, base figure or equity term is published. The listing mentions only a “Performance-Based Variable Bonus”, and the official Lever data carries no salary range. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary or equity.[1][2]
Location passes on employer evidence: Paris, hybrid, with up to two remote days a week.[1]
AI centrality passes: “Design, implement, and maintain scalable machine learning pipelines, emphasizing Large Language Models (LLMs) and generative architectures” and “Fine-tune custom LLMs for specific use cases for cost and performance optimizations”. Part of the role is internal advocacy, training and tech scouting.[1]
Seniority passes (Senior; 6+ years in software development and 3+ years deploying ML models). New to this site’s history; created April 20, 2026 per the official Lever data. The application form rendered on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Veepee runs online flash sales with over 7,000 brands. The role builds generative-AI systems for personalization and search, pricing mechanics, fraud detection and sales and media production, from problem framing and possible model fine-tuning to production deployment and monitoring.[1]
Company researchEmployer-stated on the listing: turnover of 3.3 billion euros including VAT in 2024 and 5,000 staff across 10 countries. Funding and profitability were not publicly verified in this run.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location Paris, workplace type hybrid, created April 20, 2026, no salary range.
Wanted
Doctolib
Senior Staff Machine Learning Engineer - Clinical - AI Teams (x/f/m)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris”, location type “Hybrid”; job details: “Hybrid work mode (3 days/week in the office)”.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range is published and the official Ashby data has no compensation tier; the application form asks for the applicant’s salary expectation instead. The listing names a long-term “employee value sharing plan called DoctoGrowth”, which is not stated to be equity. Pay is unknown.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary, and DoctoGrowth is not described as equity.[3][4]
Location passes on employer evidence: Paris, three office days a week.[3]
AI centrality passes: “Lead design and implementation of LLM-powered clinical agents (consultation summarization, clinical coding, evidence-grounded recommendations)” and build evaluation frameworks and production infrastructure with guardrails.[3]
Seniority passes but is demanding: 10+ years in ML/AI with 3+ years at Staff+ or Principal level, and a PhD or equivalent research experience. New to this site’s history; first published October 29, 2025 per the official Ashby data, so it has been open about a year. The application form rendered on the check date.[3][4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: the Clinical team in Doctolib’s AI and Machine Learning department builds AI systems for clinical decision-making, from voice-powered consultation assistants to LLM evaluation frameworks for clinical safety, under GDPR, MDR and EU AI Act constraints.[3]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Hybrid — Official listing header: “Paris”, “Engineering / Full-Time / Hybrid”; benefits: “A flexible remote work policy, allowing up to three days of remote work per week (depending on your team).” The office-day rule for this team is not stated.[5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range, equity or package term is published, and the official Lever data carries no salary range. Pay is unknown.[5][6]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary.[5][6]
Location is not established: the remote allowance is “up to three days” and depends on the team, so this role’s attendance rule is not explicit.[5]
AI centrality passes: “Develop NLP models … to assist with information search and recommendation within legal documents”, maintain them in production, and “Design conversational agents to accelerate lawyers’ work”.[5]
Seniority is “confirmed/senior”; the listing requires proficiency in French because the work uses French legal data. New to this site’s history; created January 23, 2025 per the official Lever data. The application form rendered on the check date.[5][6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Doctrine builds a legal AI platform used by 27,000 legal professionals in five countries. This squad covers secondary legal sources and private documents, with NLP search and recommendation models and conversational agents for drafting.[5]
Company researchEmployer-stated on the listing: a team of around 250 people. Funding, revenue and profitability were not publicly verified in this run.[5]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location Paris, workplace type hybrid, created January 23, 2025, no salary range.
Wanted
Uncovr
Senior AI Engineer
Rejected
Report date
Evidence checked
Posted
Work location
On-site — Official listing: location “Paris”, location type “On-site”. No hybrid or remote option is stated.[7]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published, and the official Ashby data has no compensation tier. A third-party job site shows a range, but it is not employer-published and was not used. Pay is unknown.[7][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location fails the configured rule: the role is listed as on-site in Paris, not hybrid with a stated attendance rule or fully remote.[7]
Pay is unpublished.[7][8]
AI centrality passes: “Build, test, and improve AI systems that support Uncovr’s product” and “Integrate language models, multimodal models, and AI services into product and backend systems”, with evaluation workflows for clinical scenarios.[7]
New requisition, first published October 26, 2026 per the ATS; the listing states no minimum years of experience. The application form rendered on the check date.[7][8]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Uncovr builds AI-powered infrastructure for surgery, turning surgical video and clinical workflow data into structured insights for surgical teams.[7]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location Paris, workplace type OnSite, published October 26, 2026, no compensation tier.
Wanted
Winamax
Senior ML / AI Engineer - Team Data (H/F)
Rejected
Report date
Evidence checked
Posted
Work location
On-site — Official listing header: “Paris”, “Permanent Contract / On-Site”, with a four-day work week. The application form asks whether a schedule of “4 jours par semaine, de 9h à 19h, en présentiel uniquement” suits the applicant.[9]
Salary evidence
No published figures
Equity
Unknown
Pay note“Rémunération : selon le profil + participation et intéressement”. No figure is published, and the official Lever data carries no salary range. Pay is unknown.[9][10]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location fails the configured rule: the role is on-site only (four days a week in the office), not hybrid or remote.[9]
Pay is unpublished (“selon le profil”).[9][10]
AI centrality passes: “Concevoir et maintenir des modèles de machine learning / IA générative”, with example projects including a chatbot and internal RAG.[9]
Seniority is in the title only; no years of experience are stated. Created March 20, 2025 per the official Lever data. The application form rendered on the check date.[9][10]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Winamax runs online poker and sports betting in France and other European markets. The Data team’s ML/AI engineer builds machine-learning and generative-AI models for business problems such as game integrity, addiction analysis and internal tools.[9]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location Paris, workplace type onsite, created March 20, 2025, no salary range.
Wanted
Mirakl
Principal AI Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: “This is a permanent position (CDI) based in Paris or Bordeaux, with 4 days on-site per week.” The application form asks whether the applicant can come to the Paris or Bordeaux offices four days per week.[1]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay note“Annual gross salary range (package): €90,000 – €100,000”. The range is labelled a package, so base salary is not confirmed, and no equity or BSPCE term is published. Even read as base pay, the €90,000 floor is not above €100,000 and has no stated equity, so it fails the pay rule.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Pay gate fails as published: the €90,000 lower bound is a “package” figure, is not above €100,000, and comes with no stated equity.[1]
Location passes on employer evidence: Paris (or Bordeaux) with four office days per week, stated on the listing and asked again on the application form.[1]
AI centrality passes: the role is the AI technical lead for one business domain and is asked to “Design and build production-grade AI systems” and to “own the agent product roadmap for your domain”. Part of the job is strategy, adoption and mentoring rather than coding.[1][2]
Seniority passes (Principal; 7+ years). New to this site’s history; first published September 7, 2026 per the official Greenhouse data. The application form rendered on the check date. A Lead AI Engineer requisition (6116425004) on the same board carries the same text and the same €90,000–€100,000 package and is not listed separately.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Mirakl sells an enterprise marketplace platform, AI-powered multichannel selling (Mirakl Connect), retail media (Mirakl Ads) and an agentic commerce infrastructure (Mirakl Nexus). This role builds AI agents for internal business domains such as Sales, GTM and Marketing or Finance & Legal, using LangGraph, LangChain, LlamaIndex, MCP and vector databases.[1]
Company researchEmployer-stated on the listing: founded in 2012, member of French Tech Next40, 750+ employees in 9 offices, 450+ marketplaces. Funding, revenue and profitability were not publicly verified in this run.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: requisition 6184130004, location Paris, France, first published September 7, 2026; a Bordeaux twin (6184132004) and Lead AI engineer requisitions 6116425004/6117258004 carry the same €90,000–€100,000 package text.
Wanted
Kpler
Senior AI Engineer (AI Enablement)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing header: “Paris / Athens”, “Engineering – Engineering Cloud Platform /Full-Time /Hybrid”. No office-day rule is published.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range, equity or package term is published on the listing, and the official Lever data carries no salary range. Pay is unknown, not below the threshold.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary.[3][4]
Location is not established: Paris is listed and the role is hybrid, but no attendance rule is published.[3]
AI centrality passes: “Design, build, and operate AI systems for internal functions — agents, assistants, and automations — owning them end-to-end from design through production operation.” The work is internal AI enablement rather than Kpler’s customer product.[3]
Seniority passes (Senior individual contributor; 5+ years). New to this site’s history; created August 25, 2026 per the official Lever data. The application form rendered on the check date.[3][4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Kpler provides trade intelligence for commodities, energy and maritime markets. This role sits in a new AI Enablement crew that builds agents, shared frameworks, knowledge-base pipelines and evaluation tooling for Kpler’s internal functions.[3]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign, build, and operate AI systems for internal functions — agents, assistants, and automations — owning them end-to-end from design through production operation.
Evidence noteMetadata summary: locations Paris and Athens, workplace type hybrid, created August 25, 2026, no salary range.
Wanted
Nebius
Senior ML Engineer (Agentic Search)
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Official listing: “Amsterdam, Netherlands; London, United Kingdom; Remote - Europe; Zurich, Switzerland”. France is not named, and the listing adds “Applicants must be authorized to work in the country in which they apply”.[5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range is published; the listing says only “Competitive compensation”, and the official Greenhouse data carries no pay range. Pay is unknown.[5][6]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary.[5][6]
Location is not established: “Remote - Europe” does not name France or say how remote hiring in France is handled.[5]
AI centrality passes: “Train and deploy ML models for retrieval, reranking, and search relevance, such as embedding models and cross-encoders, in production”, for an agent-native search platform.[5]
Seniority passes (Senior; 5+ years). New to this site’s history; first published October 16, 2026 per the official Greenhouse data. The application form rendered on the check date.[5][6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Nebius runs a full-stack AI cloud. This team builds an agent-native search platform, “the emerging web access layer for AI”, where AI systems plan, retrieve and refine results; the role trains retrieval, reranking and crawling models.[5]
Company researchEmployer-stated on the listing: listed on Nasdaq (NBIS), headquartered in Amsterdam, 1,500+ employees. Filed financial results were not read in this run.[5]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: requisition 5008613101, first published October 16, 2026, locations Amsterdam, London, Remote - Europe and Zurich, no pay range.
Wanted
Supabase
Software Engineer - AI Tooling Infrastructure
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “Remote, Global”, location type Remote; “We hire globally. We believe you can do your best work from anywhere.” France is not named.[7]
Salary evidence
No published figures
Equity
Yes
Pay noteNo salary range is published and the official ATS data has no compensation tier. The listing states “Every team member receives ESOP (equity ownership) in the company.” Pay is unknown.[7][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-engineering gate fails on the employer’s own wording: “It is a systems and infrastructure role rather than a product AI role, and it is not an SRE role.” The duties are running, securing and scaling the services behind Supabase’s AI products, including its MCP server.[7]
Pay is unpublished; equity (ESOP) is stated.[7][8]
Seniority is not stated in the title (5+ years required). New requisition, published October 31, 2026 per the ATS. The application form rendered on the check date.[7][8]
Evidence noteMetadata summary: location Remote, Global; workplace type Remote; published October 31, 2026; no compensation tier.
Wanted
GitGuardian
Senior Software Engineer, AI
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: “ENGINEERING · PARIS · HYBRID” and “Remote policy: hybrid (3 days/week at the office in Paris)”.[1]
Salary evidence
No published figures
Equity
Yes
Pay noteNo salary range is published. The listing states “Package that includes stock-options”. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual cash-base lower bound, so the role cannot be ranked under the pay rule.[1]
Location passes on employer evidence: Paris, hybrid, three office days per week.[1]
AI centrality is mixed: the Incidents team runs “an LLM risk score that rates every new incident and an MCP server that lets AI agents work with GitGuardian”, and one listed mission is “Bring AI to production in our product”. Other listed missions centre on backend performance, Django/Celery delivery and on-call support, and production LLM experience is listed as a plus rather than a requirement.[1]
Seniority passes on the title (Senior; 5+ years of web software development). New to this site’s history. Posted October 27, 2026 per the structured data on the employer’s page. The application form rendered on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: GitGuardian sells secrets detection and non-human-identity security. The Incidents team handles prioritisation, routing and remediation after a leaked credential is detected, at millions of incidents per day.[1]
Company researchEmployer-stated on the listing: Series C raised in early 2026; offices in Paris, New York City and Boston. Amount, revenue and headcount were not publicly verified in this run.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteWe put AI into production where it matters, with an LLM risk score that rates every new incident and an MCP server that lets AI agents work with GitGuardian.
Evidence noteSummary: the application form with CV upload and required personal fields rendered on the check date.
Wanted
Remanence
Member of Technical Staff — Environments / Evals
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Paris; London”, location type “Hybrid”; “A flexible hybrid setup, with a preference for working together in person.” No office-day pattern is stated.[3]
Salary evidence
No published figures
Equity
Yes
Pay noteThe employer listing says only “Competitive compensation and equity”, and the official ATS data carries no compensation. A third-party job site shows €90,000–€130,000, but that figure is not on the employer’s board and was not counted.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary. The €90K–€130K figure seen on a third-party site cannot be matched to the employer’s board.[3][4]
Location is not established: Paris is listed and the role is hybrid, but no attendance rule is published (“a preference for working together in person”).[3]
AI centrality passes: the role builds task generators, curricula, learned simulators and evaluation harnesses for training AI agents, e.g. “Build reproducible evaluation harnesses and held-out tests that detect reward exploitation, simulator shortcuts, contamination, and failures to generalize.”[3]
Seniority is not stated: the title is Member of Technical Staff with no level or years requirement. New to this site’s history; first published September 22, 2026 per the ATS. The application form rendered on the check date.[3][4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-stated: Remanence turns enterprise workflows into simulated, interactive environments where AI agents are trained and evaluated, aiming at specialised models for long-horizon tasks.[3]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteBuild reproducible evaluation harnesses and held-out tests that detect reward exploitation, simulator shortcuts, contamination, and failures to generalize.
Evidence noteMetadata summary: location Paris with London secondary, workplace type Hybrid, published September 22, 2026, no compensation component.
Wanted
n8n
Forward Deployed Engineer - EMEA
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official location list: “Berlin Office; Albania; Austria; Belgium; Bosnia; Bulgaria; Croatia; Czech Republic; Denmark; Estonia; Finland; Greece; Hungary; Ireland; Latvia; Lithuania; London Office; Montenegro; Netherlands; Norway; Poland; Portugal; Serbia; Slovakia; Slovenia; Spain; United Kingdom”, location type Remote. France is not listed.[5][6]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range is published; the listing says “Competitive compensation 💸 – We offer fair and attractive pay.” The ATS data carries no compensation tier. Pay is unknown.[5][6]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Location gate fails: France is absent from the 27-location list.[5][6]
AI-central gate fails: the main duty is enterprise workflow-automation delivery, “embedding on-site, solving real customer problems, and building reusable automation patterns”, including 5-day on-site deployment engagements. Building AI systems is not named as a primary duty.[5]
Pay is unpublished. New requisition, published October 30, 2026 per the ATS, with an application deadline of December 8, 2026. The application form rendered on the check date.[5][6]
Evidence noteEmbed on-site with enterprise customers for focused 5-day deployment engagements, arriving with a hypothesis and leaving with production workflows in place.
Evidence noteMetadata summary: primary location Berlin Office plus 26 secondary locations, none of them France; workplace type Remote; published October 30, 2026; no compensation tier.
Wanted
Vanta
Staff Software Engineer, AI and Data Risk
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “Remote - Ireland; Dublin, Ireland”; “This role is remote-first. For engineers based in the London area, we ask for 2–3 days per week in the office. For engineers based elsewhere in the UK/Ireland, we expect occasional travel to London”.[7]
Salary evidence
No published figures
Equity
Yes
Pay noteNo range is published; the listing says “Industry-competitive salary and equity”. Pay is unknown.[7]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Location gate fails: eligibility is described for the UK and Ireland only; France is not mentioned.[7]
AI centrality is not established: the role drives “privacy, AI, and risk management features” for a compliance product, built with TypeScript, React and Node.js; building or evaluating AI systems is not stated as the primary duty.[7]
Pay is unpublished. The requisition first appeared in this site’s sweep with a September 18 date and was republished October 30, 2026 per the ATS; it was not reported before. The application form rendered on the check date.[7][8]
Evidence noteThis role is remote-first. For engineers based in the London area, we ask for 2–3 days per week in the office. For engineers based elsewhere in the UK/Ireland, we expect occasional travel to London (roughly once per quarter, <20% of your time).
Evidence noteMetadata summary: location Remote - Ireland with Dublin secondary; workplace type Remote; published October 30, 2026; no compensation tier.
Wanted
Mistral AI
AI Scientist - Domain Expert Crashworthiness
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “Paris; Amsterdam; Berlin; Linz; London; Munich”, location type “Remote”.[9]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the ATS data. Pay is unknown.[9][10]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-engineering gate fails: this is a domain-expert role whose core requirement is industrial crash simulation, e.g. “Design and run high-fidelity simulation campaigns using structural mechanics solvers such as Abaqus, LS-DYNA, Ansys Mechanical, Radioss or equivalent tools.” Model training is done “with research teams”.[9]
Pay is unpublished.[9][10]
New requisition, published October 30, 2026 per the ATS. The application form rendered on the check date.[9][10]
Evidence noteDesign and run high-fidelity simulation campaigns using structural mechanics solvers such as Abaqus, LS-DYNA, Ansys Mechanical, Radioss or equivalent tools.
Evidence noteMetadata summary: location Paris with Amsterdam, Berlin, Linz, London and Munich secondary; workplace type Remote; published October 30, 2026; no compensation tier.
Remote — Official location list: “Berlin Office; Albania; Austria; Belgium; Bosnia; Bulgaria; Croatia; Czech Republic; Denmark; Estonia; Finland; Germany; Greece; Hungary; Ireland; Kosovo; Latvia; Lithuania; London Office; Moldova; Montenegro; Netherlands; Norway; Poland; Portugal; Romania; Serbia; Slovakia; Slovenia; Spain; United Kingdom”, location type Remote. France is not listed.[1][2]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay note“Western Europe / Northern Europe €159.4K – €175.3K • Offers Equity”. This is a regional tier for the listed countries; France is not among them, so no French offer range is published.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location gate fails: the role-specific location list names 31 European locations but not France. The same employer lists France on other postings on its board, such as Senior Product Manager (Enterprise), so the omission here is specific to this role.[1][2]
AI centrality passes on the listing: “This is a Staff individual contributor role in Developer Platform and the technical driving force behind ACDC (Agent-centric Development Cycle), our engineering-wide initiative for agentic development.”[1]
Recorded as a near-miss on October 17. Material changes since then: the ATS publish date is now October 29, 2026; the description now names a Staff individual contributor level; the Western/Northern Europe tier rose from €114.3K–€149.3K to €159.4K–€175.3K. The location list still omits France, so the role is rejected rather than reconsidered.[1][2]
The same-page application form rendered on the check date.[1]
Evidence noteThis is a Staff individual contributor role in Developer Platform and the technical driving force behind ACDC (Agent-centric Development Cycle), our engineering-wide initiative for agentic development.
Evidence noteMetadata summary: primary location Berlin Office plus 30 secondary locations, none of them France; workplace type Remote; published October 29, 2026; Western Europe / Northern Europe salary tier €159,400–€175,300 per year with equity.
Wanted
Filigran
Staff AppSec / Product Security Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official listing: location “France”, location type “Remote”; the profile asks for someone “Comfortable working in a remote, async-first environment”.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity term is published on the listing or in the ATS compensation data. Pay is unknown, not below the threshold.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-central gate fails: the primary duty is product security across the XTM platform, for example “Drive secure SDLC and supply chain tooling in CI (SAST, SCA, secret and container scanning, Renovate, Snyk, SBOMs, signed releases), tuned so teams trust the results.”[3]
AI appears as one review area and as a productivity tool, not the core assignment: “Lead threat modeling and security reviews in feature refinements with Product Owners: authentication and RBAC, multi-tenancy, connectors, the GraphQL API, and AI and agent features.”[3]
Location and seniority pass on the listing (France, remote; Staff). Pay gate unverified: no published salary.[3]
New to this site’s history; published October 29, 2026 per the ATS. The same-page application form rendered on the check date.[3]
Evidence noteLead threat modeling and security reviews in feature refinements with Product Owners: authentication and RBAC, multi-tenancy, connectors, the GraphQL API, and AI and agent features.
Wanted
Langfuse (ClickHouse)
Langfuse - Senior Cloud Infra Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Official listing: location “Germany; France; Italy; Portugal; Spain; The Netherlands”, location type “Hybrid”. The listing does not state an office or attendance pattern for France-based hires.[4][5]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“Salary €90K – €160K • Offers Equity”; the official ATS data classifies the €90,000–€160,000 component as an annual EUR salary. The listing adds that every new team member receives stock options.[4][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralfail
Paypass
Product researchunknown
Gate notes
AI-central gate fails: this is a cloud infrastructure and SRE role for an LLM-observability product. The listing says “You'll own uptime, performance, and cost efficiency across our entire cloud footprint” on AWS ECS Fargate and ClickHouse Cloud.[4]
Pay passes on employer evidence (€90,000 annual salary floor with equity). Seniority passes (Senior).[4][5]
Location is not established: France is listed, but the type is Hybrid and no French office or attendance rule is given. A remote-first statement quoted from the Langfuse careers page on October 20 was not found on that page today.[4]
New to this site’s history; published October 29, 2026 per the ATS. Langfuse SDK, Product and Backend roles appear in the October 20 report and are separate requisitions.[5]
Evidence noteMetadata summary: primary location Germany with France, Italy, Portugal, Spain and The Netherlands as secondary locations; workplace type Hybrid; published October 29, 2026; salary component €90,000–€160,000 EUR, interval 1 year, plus equity.
Wanted
GitLab
Staff Solutions Architect, AI Subject Matter Expert
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Official Greenhouse locations: “Remote, France; Remote, Germany; Remote, Netherlands; Remote, Spain; Remote, United Kingdom”.[1][2]
Salary evidence
No published figures
Equity
Yes
Pay noteNo salary range is published on the listing, and the Greenhouse pay-transparency field is empty. The benefits list names “Equity Compensation & Employee Stock Purchase Plan”. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-central hands-on engineering gate fails: this is a field pre-sales specialist role. The listing says “Your success will be measured by technical win rate and bookings on SME-engaged opportunities” and asks for “Substantial experience in technical pre-sales, solutions architecture, consulting, or software engineering”.[1]
AI is one of four possible specialisms, not the guaranteed assignment: “You'll specialize in one product group, not all four.” The AI track reads: “In AI & Competitive, you'll focus on GitLab Duo Agent Platform, agentic workflows, AI-native development, competitive displacement including GitHub, and migrations.”[1]
Location passes on employer evidence (remote, France listed). Pay gate unverified: no published base salary.[1][2]
New to this site’s history. First published October 16, 2026 per Greenhouse; the board record was updated October 28, which is when the daily sweep first matched it.[2]
Evidence noteMetadata summary: locations Remote, France; Remote, Germany; Remote, Netherlands; Remote, Spain; Remote, United Kingdom; first published October 16, 2026; updated October 28, 2026; no pay-transparency ranges.
Wanted
Parallel
Forward Deployed Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — ATS location: Paris, workplace type Hybrid. The listing says: “Flexibility: on average one day of remote work per week, and one remote week per year.”[3][4]
Salary evidence
No published figures
Equity
Yes
Pay noteNo figure is published; the listing offers a “Competitive salary and equity package”. Pay is unknown, not below the threshold.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-central gate fails: the primary duty is field deployment and integration. The mission is to “Technically lead end-to-end deployments by understanding each facility's IT infrastructure, adapt our platform to their constraints (interfaces, integrations, security, network architecture, on-prem deployments), and unblock projects by getting hands-on with the product when needed.”[3]
AI is a secondary requirement: “You've played with LLMs and AI agents — whether on side projects or professionally (LangChain, Hugging Face)”. The product itself is AI-based: “Our technology combines Large Language Models with system-level automation to handle real work inside existing hospital tools.”[3]
Seniority is not stated as senior or staff (the listing asks for 5+ years of full-stack experience), and pay is unpublished.[3]
Not new: published September 7, 2026 per the ATS. The posting was on the board from October 20 to 23, missing from this site’s sweeps from October 24 to 28, and listed again today. It was never recorded in an earlier report.[4]
Evidence noteTechnically lead end-to-end deployments by understanding each facility's IT infrastructure, adapt our platform to their constraints (interfaces, integrations, security, network architecture, on-prem deployments), and unblock projects by getting hands-on with the product when needed.
Pay note“€55–80k/year + equity”. The page does not say whether the range is base salary or a package. Either way the €55,000 floor is below €80,000.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published floor of €55,000 is below the €80,000 equity branch and the above-€100,000 cash branch.[5]
Location passes: Paris with a stated 50% on-site attendance pattern.[5]
AI centrality passes on the listing: “Build production-grade agents in-house on the Idun Platform: from architecture to eval to deployment, for our consulting clients.” The work is delivered for consulting clients.[5]
Found through web search and confirmed on the employer’s own careers page. Posted October 7, 2026 per the employer page; new to this site’s history.[5]
Remote — “Location: France, Germany, Spain, or Italy — remote within these hiring locations.” ATS location: Paris, France, workplace type Remote.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing, and the official ATS data carries no compensation tiers for this role. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual base salary and no equity terms, so neither branch of the pay rule can be checked.[1][2]
Location passes on employer evidence: remote work is open within France, Germany, Spain or Italy.[1]
AI centrality passes: “Design and ship agentic tools and AI-powered workflows that gather context from multiple internal systems, orchestrate models and tools, parse structured outputs, and handle uncertainty and failure safely”. The listing adds: “This is not a conventional backend role using AI only as a coding assistant, nor an ML research role.”[1]
Seniority passes: a staff-level role that asks for “Staff-level autonomy and judgement”.[1]
Same requisition this site rejected on October 12 for unpublished pay. The employer republished it on October 27 with a detailed team description; pay is still unpublished, so the earlier decision stands. It is listed again only because of the republication. The application form was not opened because the pay gate is unverified. The October 12 record is preserved in that day’s archived report.[2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Qonto is a business finance workspace with banking at its core for SMEs. The role sits in the AI Compliance Tooling team within Financial Crime Compliance. The listing says the team does not build fraud-detection or KYC/KYB rule engines; it builds the AI automation layer used by human investigators. One current system “gathers information from 10–15 internal tools and produces a structured compliance analysis for a person to approve or edit”, cutting treatment time from about 15–20 minutes to about 30 seconds (employer claim). A separate product source was not read today, so the product gate is not marked as passed.[1]
Company researchNot researched today because the role was not ranked. The listing states 1,700 staff, over 750,000 customers in 8 European countries and profitability since 2023; these are employer statements, not verified financial disclosures.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: published October 27, 2026; location Paris, France; workplace type Remote; no compensation tiers.
Wanted
Mendo
Senior AI Engineer - AI Tech Lead
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — “Location: Paris, Châtelet – Hybrid (2 days on-site / 3 days remote)”.[3]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay note“Compensation: €75,000–€85,000 gross annual package (base + 12% target variable) + BSPCE (equity), depending on step”. The figures are a package including variable pay, so the base salary is lower than the quoted floor.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published floor is a €75,000 gross package that includes a 12% target variable. That is below €80,000 even before removing the variable part, so the equity branch fails, and the cash-only branch (above €100,000 base) fails too.[3]
Location passes: Paris hybrid with an explicit two-days-on-site pattern.[3]
AI centrality and seniority pass on the listing: “Train and ship on-device AI models, improve production features, and help leading organizations turn generative AI into everyday value”, and it asks for “6+ years in ML / AI engineering”.[3]
New requisition, posted October 23, 2026 per the employer page. It is separate from Mendo’s earlier Senior AI Engineer listing, which this site recorded as unavailable on October 15 and which is no longer on the board.[3][4]
Evidence noteMetadata summary: the board lists Senior AI Engineer - AI Tech Lead (8611034); the earlier Senior AI Engineer listing (6979783) is not listed.
Wanted
Docker
Principal Forward Deployed Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Remote — ATS locations: England, Portugal, Spain, Italy and Germany; workplace type Remote. France is not listed. The listing also says: “Travel to customer sites to build in person with customer teams, estimated at up to 50%.”[5][6]
Salary evidence
No published figures
Equity
Yes
Pay noteThe only published range is “United States: €121K – €198K + equity”. No range is published for any European location, and no range applies to France.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location gate fails: the official ATS locations are England, Portugal, Spain, Italy and Germany, with no France eligibility stated.[5][6]
Pay is unverified for Europe: the only published figure is for the United States.[5]
AI centrality passes on the listing: deliverables include “custom MCP servers with tightly integrated Docker Sandbox environments for coding agents and containerized agents”, and it requires hands-on production experience with LLMs, agents, MCP and agent evaluation.[5]
New to this site’s history; first published October 27, 2026 per the ATS. A large share of the role is customer-facing delivery with up to 50% travel.[5][6]
Evidence noteDeliver working technical artifacts to customers: custom MCP servers with tightly integrated Docker Sandbox environments for coding agents and containerized agents
Evidence noteMetadata summary: primary location England; secondary locations Portugal, Spain, Italy, Germany; remote; published October 27, 2026; no compensation tier summary.
Wanted
Shift Technology
Lead AI Engineer - ML/LLM Ops & Agentic AI
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — ATS location: “France - Paris; France - Remote”. Benefits: “Flexible remote and hybrid working options”. The process ends with “A final round at our office with multiple interviews”. No weekly attendance pattern is stated.[1][2]
Salary evidence
No published figures
Equity
Unknown
Pay note“Competitive Salary and a variable component tied to personal and company performance”. No figure or equity terms are published on the listing, and the official ATS data carries no pay range. Pay is unknown, not below the threshold.[1][2]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: the employer publishes no annual base salary and no equity, so neither branch of the pay rule can be checked.[1][2]
Location passes on employer evidence: the ATS lists “France - Remote” alongside Paris, and the benefits mention remote and hybrid options.[1][2]
AI centrality passes: the role is to “design and ship production-grade agentic workflows (LLM orchestration, tool calling, MCP, RAG, structured outputs)”, own LLMOps standards (evaluation harnesses, prompt and agent versioning, tracing, cost governance, release gates) and own the inference layer for self-hosted open-weight models.[1]
Seniority passes: a hands-on lead role that is to “coach and develop 2 to 5 people” while continuing to build.[1]
New requisition, first published October 26, 2026 per the ATS. It is separate from Shift’s Staff AI Engineer near-miss of October 23, which is still listed with no change in the screened fields. The Greenhouse application form was present on the listing page on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: “Shift delivers AI agents that transform insurers' most critical work.” The role sits in Shift’s Data & AI Platform Engineering team. Named duties include a Databricks-based ML platform (Unity Catalog, MLflow, serving on AKS), self-hosting open-weight foundation models with Cloud Ops, and time-boxed proofs of concept such as relational foundation models for claims fraud detection and agentic ingestion of medical billing sources. English is required; French is a plus. A separate company or product source was not read today, so the product gate is not marked as passed.[1]
Company researchNot researched today because the role was not ranked. Shift’s funding, financial condition and headcount are not publicly verified in this report.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location “France - Paris; France - Remote”; first published October 26, 2026; requisition 950; no pay range in the pay-transparency field.
Remote — ATS location: “Anywhere in France, Belgium, Spain”. “Location: you must be legally eligible to work in France, Belgium, or Spain.” “Remote work: we offer remote work flexibility, but we value in-person collaboration.”[3][4]
Salary evidence
No published figures
Equity
Unknown
Pay note“Generous equity packages complement your base salary.” No salary figure is published on the listing, and the official ATS data carries no compensation tier for this role. Pay is unknown, not below the threshold.[3][4]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual base salary for this role.[3][4]
Seniority not established: the title carries no senior or staff level, and the listing asks for “3+ years in software engineering, with solid backend fundamentals”.[3]
AI centrality passes: the Agentic Platform area says “We own the environment agents run in (infrastructure, runtime, evaluation, observability, skills lifecycle)”, on a stack that includes PydanticAI, Temporal, MCP and sandboxes.[3]
Location passes on employer evidence: open anywhere in France for people legally eligible to work there, with remote flexibility.[3][4]
New to this site’s history; first published October 26, 2026 per the ATS. The application form was not opened because the role failed two gates.[4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Alan sells health insurance with prevention and care services. The listing says agents already answer members’ health questions, resolve support requests, read medical documents and detect fraud, and that “In 2027, we expect to run more AI agents than we have employees.” The Agentic Platform area builds the shared runtime, evaluation and observability layer, plus the internal assistant Hopper and an Agent Studio. Company-reported figures in the listing (40K+ companies, 1M+ members, €800M+ ARR, 1000+ people) were not independently verified.[3]
Company researchNot researched today because the role was not ranked. The listing’s company figures are employer claims, not verified financial disclosures.
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: location “Anywhere in France, Belgium, Spain”; published October 26, 2026; no compensation tiers.
Wanted
Dust
Senior AI Support Engineer (Paris)
Rejected
Report date
Evidence checked
Posted
Work location
On-site — ATS location: Paris, workplace type On-site. “We're prioritizing building our team with an in-person culture at our offices in Paris, London, San Francisco, and New York”.[5][6]
Salary evidence
Known annual EUR base
Equity
Yes
Pay noteOfficial ATS data: “€50K – €100K”, a one-year EUR Salary component titled “Depending on experience”. Benefits: “Significant equity package in a Sequoia-backed startup”. This site’s October 17 report recorded no published salary figure for this requisition.[5][6]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published annual salary floor is €50,000, below €80,000 even with equity and well below the €100,000 cash-only branch.[5][6]
Changed since this site’s October 17 near-miss: the employer republished the requisition on October 23 with a published range. The new evidence settles the pay gate as a fail.[6]
The role is a senior support-engineering position (“As a Senior AI Support Engineer, you will help define what world-class support looks like in an AI-native company.”), recorded as AI-relevant on October 17; that decision is not revisited because pay fails.[5]
Evidence noteMetadata summary: Salary component €50,000–€100,000, interval 1 YEAR, currency EUR, tier “Depending on experience”; published October 23, 2026; workplace type On-site.
Wanted
Photoroom
Product ML Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — ATS location: Paris. Benefits: “Hybrid working from our Paris headquarters”. No attendance pattern is stated.[7]
Salary evidence
No published figures
Equity
Unknown
Pay note“Competitive salary + meaningful equity”. No salary figure is published on the listing or in the official ATS data.[7][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Seniority gate fails: the title has no senior or staff level and the listing asks for “You have 2+ years of hands-on experience building and shipping ML-powered products”.[7]
Pay is unpublished and the hybrid attendance pattern is unstated.[7][8]
Same title and content as the Photoroom listing this site rejected on October 13; the employer reposted it under a new ID on October 26 and the old ID is no longer on the board. No material change was found.[8]
Evidence noteMetadata summary: published October 26, 2026; location Paris; no compensation tiers; previous Product ML Engineer ID absent from the board.
Wanted
lemlist
Senior Applied AI Engineer
Rejected
Report date
Evidence checked
Posted
Not published by the employer
Work location
Unknown — Paris. A search snippet mentioned possible full remote, but the employer page no longer shows the listing details.[11]
Salary evidence
No published figures
Equity
Unknown
Pay noteNot published on the employer page on the check date.[11]
Recorded gates (as checked on the report date)
Listing openfail
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Open-listing gate fails: the employer’s job page states “This position has been filled!”, and no application path is offered.[11]
Unknown — Paris. The job-board page no longer shows attendance details.[12]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay noteThe closed job-board page shows “80 000 - 110 000 € / an”; it is not labelled as base salary and is not employer-hosted. A search snippet mentioned equity, which could not be confirmed on any open page.[12]
Recorded gates (as checked on the report date)
Listing openfail
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Open-listing gate fails: the job-board page says the offer is no longer available, and the employer’s own careers board lists no Staff Engineer role on the check date.[12][13]
AI centrality was not established: the role was described as a staff-level engineering reference for the whole platform, with the AI layer as one part of it.[12]
Evidence noteOnly one open role was listed on the check date: Account Executive - DACH region, Paris.
Wanted
JAKALA
AI Engineer – Senior Consultant
Near-miss
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — “Bureau en plein centre de Paris” and “Télétravail possible jusqu’à 2 jours par semaine” (remote work possible up to two days a week). Lieu : Paris.[1]
Salary evidence
No published figures
Equity
Unknown
Pay note“Rémunération selon profil et expérience” (pay depends on profile and experience). The employer page publishes no salary figure and no equity. Third-party job sites show conflicting figures for this listing and were not used. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: the official listing publishes no annual base salary and no equity, so neither branch of the pay rule can be checked.[1]
Location passes on employer evidence: a central-Paris office with remote work capped at two days a week.[1]
AI centrality passes: the engineer designs and builds LLM solutions (conversational agents, RAG, tool-using agents) and ML models, keeps them running in production, and monitors token cost, model choice, caching, latency and reliability of agents in production.[1]
Seniority is only partly established: the title is Senior Consultant, but the listing asks for more than three years of experience, and the role also covers client use-case discovery and proofs of concept in a consulting setting.[1]
New to this site’s history. No posting date is published on the employer page. The application form (name, current role, contact details, CV upload) opened from the listing’s Postuler button on the check date.[1]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: JAKALA describes itself as a European data and AI consulting firm for marketing, founded in Italy in 2000, present in 20 countries with 3,000 staff, and established in France in 2023 with close to 250 consultants at the end of 2025 (employer statements, not independently verified here). This role builds LLM and ML solutions for client use cases. Separate product diligence was not advanced because the role is not ranked.[1]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteRémunération selon profil et expérience
Wanted
INOCO
AI Security Engineer
Rejected
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — Paris, “Télétravail occasionnel” (occasional remote work); workplace 6 Rue du 8 Mai 1945, 75010 Paris. No attendance pattern is stated.[2]
Salary evidence
No published figures
Equity
Unknown
Pay note“Salaire : Non spécifié” (salary not specified) on the employer’s job page. A company-wide perk list mentions a share purchase plan, BSPCE or stock options, which is not a role-specific equity offer. A higher figure shown on a third-party job site was not used.[2]
Recorded gates (as checked on the report date)
Listing openfail
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Open-listing gate fails: the employer’s job page states “Cette offre n’est plus disponible” (this offer is no longer available), and no application path is offered.[2]
The role would have been AI-central: threat modeling of LLM and agentic applications, agent permission and isolation design, guardrails, and red teaming client applications before and after release.[2]
Pay was not published by the employer, and the location wording gives no attendance pattern.[2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: INOCO describes itself as an independent, employee-owned IT services firm of about 60 people that takes client AI projects from scoping to production (employer statement, not independently verified here).[2]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Bureaudestalents (recruiter for an unnamed fintech client)
Senior AI Engineer
Rejected
Report date
Evidence checked
Posted
Not published by the employer
Work location
Hybrid — Paris 75008. “Work setup: Hybrid (3–4 days/week in the office)”.[3]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay noteJob summary: “Salaire : 85K à 120K €”. Listing body: “Salary: Depending on experience ( + stock options )”. The range is not labelled as base salary.[3]
Recorded gates (as checked on the report date)
Listing openfail
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Open-listing gate fails: both copies of this recruiter listing state “Cette offre n’est plus disponible” (this offer is no longer available), and no application path is offered.[3][4]
The role would have been AI-central: designing and deploying LLM features for document understanding, financial data extraction and semantic search, and owning the AI roadmap and evaluation frameworks.[3]
The employer is an unnamed client of a recruitment firm, described as a Paris fintech SaaS company of about 15 people that recently raised €4 million (recruiter statements), so the company could not be researched separately.[3]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: the recruiter describes the client as building proprietary financial data and analytics tools for analysts, portfolio managers and research teams.[3]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Hybrid — Location: Paris; New York. Location Type: Hybrid. “Hybrid flexibility — We're in the Heart of Paris and our team is in 3x a week!”[1]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay note“For candidates based in France, the pay range for this role is €127,000-€171,000.” The listing separately offers “Variable pay — Based on real Vibe business goals.” The range is not labelled as base salary, no equity is stated for this role, and the official ATS data carries no structured compensation for it.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unconfirmed: the France pay range starts at €127,000, above the €100,000 floor, but the listing does not say whether the range is annual gross base salary or includes the separately listed variable pay. The role is held as a near-miss rather than ranked.[1]
Supporting but not decisive: for a sibling Vibe listing that uses the same “pay range” wording, Vibe’s official ATS data codes the range as a salary component separate from bonus and equity. That is evidence about Vibe’s convention, not a statement for this role.[3]
Location passes on employer evidence: Paris hybrid, with the team in the office three times a week.[1]
AI centrality passes, with a caveat: the lead designs and builds the clustering and entity-resolution algorithms that attach every observed ad event to a real identity, drives accuracy improvements through evaluation frameworks, and feeds the training data Vibe’s models learn from. It is a player-coach role that also manages one senior engineer.[1]
Seniority passes: the listing asks for 8+ years building and operating production ML or data systems, including technical leadership as a tech lead, staff engineer or team lead. First published August 6, 2026 and new to this site’s history. The application form rendered on the check date.[1][2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Vibe runs a streaming-TV advertising platform that, it says, more than 12,000 brands use to reach 120 million+ households, and states it became a Walmart company in August 2026. This role leads the three-person Identity squad on the Data Platform team, which resolves identity at US scale and is extending it into audience construction and attribution. Separate product diligence was not advanced because the role is not ranked.[1]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteFor candidates based in France, the pay range for this role is €127,000-€171,000. Design and build clustering algorithms and pipelines that resolve identity at US scale
Evidence noteThe pay range for this role is €99,000 - €171,000.
Wanted
White Circle
Research Scientist/Engineer (Agentic Systems)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris; London. Location Type: Hybrid. “Office in central London/Paris with flexible hybrid setup”; no number of office days is stated.[4]
Salary evidence
No published figures
Equity
Yes
Pay note“Competitive compensation, including equity.” No salary figure is published on the listing, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: equity is offered, but no employer-published annual cash-base lower bound.[4]
Location gate unconfirmed: Paris hybrid is stated, but the explicit attendance pattern the rule requires is not.[4]
AI centrality passes: the role builds adversarial single- and multi-agent environments that push LLM agents to failure, runs experiments against external APIs and White Circle’s own models, and catalogues agent failure modes with tooling to surface them at scale.[4]
No level is named; the listing asks for a track record of empirical research in agents, red-teaming or post-training. First published June 29, 2026, it is new to this site’s history, not newly created. The application form rendered on the check date.[4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: White Circle describes itself as an AI safety company whose platform tests and enforces natural-language policies on AI systems; it says it has raised $70M in total, including a Series A, and processes over 100M API calls a month (employer statements, not independently verified here). This research team studies how agents fail and turns findings into guardrails. Separate product diligence was not advanced because the role is not ranked.[4]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteTL;DR: We're looking for a research scientist to build autonomous, large-scale environments that push LLM agents (single and multi-agent) to failure, and study how they actually break.
Wanted
White Circle
Research Engineer (Evals)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris; London. Location Type: Hybrid. “Office in central London/Paris with flexible hybrid setup”; no number of office days is stated.[5]
Salary evidence
No published figures
Equity
Yes
Pay note“Competitive compensation, including equity.” No salary figure is published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: equity is offered, but no employer-published annual cash-base lower bound.[5]
Location gate unconfirmed: Paris hybrid is stated without an attendance pattern.[5]
AI centrality passes: the engineer owns the internal benchmark suite for content and agentic guardrails, builds evals for the company’s flagship models and studies real-world LLM and agent failure modes.[5]
No level is named. First published June 28, 2026 and new to this site’s history. The application form rendered on the check date.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: White Circle describes itself as an AI safety company whose platform tests and enforces natural-language policies on AI systems; it says it has raised $70M in total, including a Series A, and processes over 100M API calls a month (employer statements, not independently verified here). This role maintains the benchmarks that measure those guardrails and models. Separate product diligence was not advanced because the role is not ranked.[5]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteOwn and maintain our internal benchmark suite, covering single/multi-turn content guardrails and agentic safety.
Wanted
Vibe
Staff or Senior Data Platform Engineer - Data Infrastructure
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris. Location Type: Hybrid. “Hybrid flexibility — We're in the Heart of Paris and our team is in 3x a week!”[3]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“The pay range for this role is €99,000 - €171,000.” Vibe’s official ATS data lists this €99K–€171K range as an annual salary component, separate from “Offers Bonus” and “Offers Equity”, and the listing offers “Equity — Employee Stock Ownership Plan.”[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Paypass
Product researchunknown
Gate notes
AI-central engineering gate fails: the role owns Kafka streaming, Dagster batch orchestration and the ClickHouse serving path for the data platform. Machine-learning teams are among its internal users, but building or operating AI systems is not a primary duty.[3]
Pay passes the equity branch on employer evidence: a €99,000 annual salary lower bound with an explicit employee stock ownership plan.[3]
Location passes: Paris hybrid, three office days a week. The application form rendered on the check date.[3]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Vibe runs a streaming-TV advertising platform that, it says, more than 12,000 brands use to reach 120 million+ households, and states it became a Walmart company in August 2026. This Data team role runs the batch, streaming and reporting stacks behind advertiser reporting and internal analytics.[3]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign and operate real-time event processing on Kafka at production scale
Wanted
Vibe
Senior SRE/ML Ops Engineer - Performance
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris. Location Type: Hybrid. “Hybrid flexibility — We're in the heart of Paris and our team is in 3x a week!”[6]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay note“For candidates in France, the pay range for this role is €70,000-€110,000.” Variable pay is listed separately; no equity is stated for this role.[6]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published range starts at €70,000, below both the €100,000 floor and the €80,000-plus-equity branch, whatever the range covers.[6]
AI centrality is partial: the engineer owns reliability, latency and cost for the prediction-serving and bidding stack and works with the data scientists who ship models into it, rather than building the models.[6]
Location passes: Paris hybrid, three office days a week. The application form rendered on the check date.[6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Vibe runs a streaming-TV advertising platform that, it says, more than 12,000 brands use to reach 120 million+ households, and states it became a Walmart company in August 2026. The Performance team owns the prediction models, bidding engine, pacing and delivery stack.[6]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteFor candidates in France, the pay range for this role is €70,000-€110,000.
Wanted
EverAI
Mid/Senior AI LLM Engineer (Full Remote - Europe)
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Location: Europe. Location Type: Remote. “Work From Anywhere: Fully remote.” The listing names no specific countries and says the company prefers B2B contracts.[7]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay note“Annual Compensation Range (Gross) €75K – €125K.” The listing says “Published ranges reflect total gross annual compensation, including any applicable target bonus.” No equity is stated.[7]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published range is total compensation including target bonus, not base salary, and its €75,000 lower bound is below both branches of the rule.[7]
AI centrality passes: the engineer fine-tunes open-weight LLMs with SFT, LoRA/QLoRA, DPO and GRPO, builds training data and evals, and ships models to production.[7]
Location is unconfirmed for France: “Europe” remote with no named eligible countries, and a stated preference for B2B contracts. The listing also says the work involves explicit adult content. First published October 20, 2026; the application form rendered on the check date.[7]
Evidence notePublished ranges reflect total gross annual compensation, including any applicable target bonus.
Wanted
Vibe
Staff or Senior Machine Learning Engineer - Identity & Audience
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris. Location Type: Hybrid. “Hybrid flexibility — We're in the Heart of Paris and our team is in 3x a week!”[1]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay note“The pay range for this role is €99,000 - €171,000.” The listing separately offers “Equity — Employee Stock Ownership Plan” and “Variable pay — Based on real Vibe business goals.” The range is not labelled as annual base salary, and the official ATS data carries no structured compensation for this role.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unconfirmed: the listing publishes a €99,000–€171,000 pay range and explicit equity, but does not say whether the range is annual gross base salary or includes variable pay. Only a confirmed base lower bound of at least €80,000 with equity would pass, so the role is held as a near-miss rather than ranked.[1]
Location passes on employer evidence: a Paris hybrid role with the team in the office three times a week.[1]
AI centrality passes: the engineer researches and builds machine-learning, clustering and statistical algorithms that resolve users and households from billions of advertising events, and turns them into production systems that feed targeting, measurement and model training sets.[1]
Seniority passes: the title is Staff or Senior, and the listing asks for 5+ years as an ML engineer or data scientist building and operating production services, plus data-scale tools such as Dagster, Spark or Iceberg.[1]
First published October 8, 2026 and new to this site’s history. The application form rendered on the check date.[2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Vibe runs a streaming-TV advertising platform used, it says, by more than 12,000 brands to reach 120 million+ households, and states it became a Walmart company in August 2026. This role sits on the Data Platform team and owns identity resolution, the layer that decides which ad reaches which household and whether it worked. Separate product diligence was not advanced because the role is not ranked.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThe pay range for this role is €99,000 - €171,000. We will support relocation to Paris for great candidates. Hybrid flexibility — We're in the Heart of Paris and our team is in 3x a week!
Staff Product Engineer - AI Core - remote friendly
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Location: Paris, France. “You can apply to work at Qonto as long as you're living in (or willing to relocate to) France, Germany, Spain, or Italy. This role is hybrid or remote.”[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing, and the official ATS data carries no compensation for this role. Pay is unknown, not below the threshold.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual cash-base lower bound.[3]
Location passes on employer evidence: applicants living in France are eligible, and the role may be done hybrid or remote.[3]
AI centrality passes, with a caveat: the engineer restructures Qonto’s internal AI Software Factory and ships agents for bug resolution, code review, QA, auto-approval and safe auto-merge, starting with autonomous bug resolution. Part of the role is acting as the product manager of that internal tool.[3]
Seniority passes on title (Staff). Requirements are strong engineering, the ability to read Python, and working knowledge of how models, prompting and evaluation work.[3]
First published October 22, 2026. The application form rendered on the check date.[3]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Qonto is a business-finance and banking platform for European SMEs; the listing says it has 1,700 employees, over 750,000 customers in 8 countries and has been profitable since 2023 (employer statements, not independently verified here). This internal role builds coding agents for Qonto’s own engineers, not a customer-facing product. Separate product diligence was not advanced because the role is not ranked.[3]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteYou can apply to work at Qonto as long as you're living in (or willing to relocate to) France, Germany, Spain, or Italy. This role is hybrid or remote.
Wanted
Mistral AI
Research Engineer - Cybersecurity (RL Environments)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris. Location Type: Hybrid. The listing states no number of office days or other attendance terms.[4]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[4]
Location gate unconfirmed: Paris hybrid is stated, but the explicit attendance pattern the rule requires is not.[4]
AI centrality passes: the engineer designs reinforcement-learning environments that simulate security scenarios, builds pipelines that generate synthetic cybersecurity training instances at scale, and runs evaluations whose results feed Mistral’s production training runs.[4]
No level is named. Requirements include hands-on offensive or defensive security expertise, Python plus reading C/C++, and Docker, Kubernetes and cloud deployment experience. First published September 25, 2026, it is new to this site’s history, not newly created. The application form rendered on the check date with Mistral’s 3-applications-in-90-days limit.[4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: part of Mistral’s Science department, this track develops its frontier models’ abilities in vulnerability discovery and remediation, security analysis and incident response, across the offensive-to-defensive spectrum. Separate product diligence was not advanced because the role is not ranked.[4]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteConcretely, that means designing and building RL environments that reproduce realistic security scenarios, pipelines that generate synthetic training instances at scale, and experiments and evaluations that show what our models can actually do. Candidates may submit up to 3 total applications within a 90 day period across all open roles.
Wanted
Shift Technology
Staff AI Engineer - Architecture skills
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Location: “Any Shift Location; France - Paris”. Benefits list “Flexible remote and hybrid working options”; no attendance pattern or remote-country eligibility is stated.[5]
Salary evidence
No published figures
Equity
Unknown
Pay noteThe listing promises “Competitive Salary and a variable component” but publishes no figure or equity terms, and the official ATS data carries no pay range. Pay is unknown, not below the threshold.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[5]
Location gate unverified: Paris is named alongside “Any Shift Location”, but neither an attendance pattern nor explicit remote eligibility for France is given.[5]
AI centrality passes: the hands-on Staff AI Engineer designs and ships agentic workflows, LLM orchestration layers and tool-calling integrations, moves AI prototypes into production microservices on Azure, and sets up evaluation, tracing and token-budget practices.[5]
Seniority passes on title (Staff). First published October 3, 2026, it is new to this site’s history. The Greenhouse application form rendered on the same page on the check date.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Shift Technology sells AI agents for insurers’ claims and fraud work; this role is the technical anchor for its AI platform capabilities, including MCP, advanced RAG patterns and structured LLM outputs. Separate product diligence was not advanced because the role is not ranked.[5]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteWe are looking for an experienced, hands-on Staff AI Engineer to serve as the technical anchor for our AI capabilities; someone who is deeply passionate about building, prototyping, and scaling modern AI/LLM systems, while helping shape our broader AI engineering strategy.
Wanted
Doctolib
Senior AI Engineer - Search & Recommendation - Patient Team (x/f/m)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris. Location Type: Hybrid. Job details: “Workplace: Levallois”. No number of office days is stated.[6]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published; the only figures are meal-voucher amounts. The listing mentions an employee value-sharing plan but no equity grant. Pay is unknown, not below the threshold.[6]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[6]
Location gate unconfirmed: the workplace is Levallois, next to Paris rather than in the city, and the hybrid attendance pattern is not stated.[6]
AI centrality passes: the engineer designs the production retrieval, ranking and reranking pipeline (vector search, semantic retrieval, LLM or managed rerankers) behind Doctolib’s patient health companion, with offline and online evaluation built into every stage.[6]
Seniority passes on title (Senior). The ATS header labels the employment type “Contract”, while the body says “Permanent position”. A near-identical sibling listing, “Senior AI Engineer - Patient Team”, carries the same text under a different requisition. Both application forms rendered on the check date.[6][7]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Doctolib’s Patient domain builds a health companion that it says helps 100M patients across Europe find the right care; this role owns the search and recommendation architecture behind it. Separate product diligence was not advanced because the role is not ranked.[6]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign and build the production search & recommendation architecture: full retrieval, ranking, reranking pipeline with standard and off-the-shelf components (vector search, semantic retrieval, LLM/managed rerankers).
Evidence noteWe are looking for a Senior AI Engineer to join the Patient team in Paris.
Wanted
ElevenLabs
Applied AI Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Location: United Kingdom; Bulgaria; Dublin; France; Germany; Hungary; Italy; Netherlands; Poland; Spain. Location Type: Remote. “This role is remote-first, so it can be executed from anywhere in Europe, with the ability to operate in GMT and GMT+1 timezones.”[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing, and the official ATS data carries no compensation for this role. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual cash-base lower bound, so the role cannot be ranked under the pay rule.[1]
Location passes on employer evidence: fully remote within Europe, with France named among the listed locations.[1]
AI centrality passes: the engineer builds agents that run parts of the business, using the company’s own agents platform, orchestration tools and frontier models, and is responsible for making them reliable, measurable and safe. Requirements include Python, SQL, LLM APIs and agent orchestration, and evaluating LLM systems beyond a demo.[1]
Seniority is not stated: the company says it does not use job titles, and the listing gives no level or years of experience. The role sits in Marketing Engineering and is forward-deployed to growth teams.[1]
The application form rendered on the check date and asks applicants to describe a shipped agent or LLM system and how it was evaluated.[2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: ElevenLabs builds AI voice models and three platforms (ElevenAgents for voice and chat agents, ElevenCreative for media generation, and ElevenAPI for developer access to its audio models). This role builds internal agents for the company’s growth teams rather than a customer-facing product. Separate product diligence was not advanced because the role is not ranked.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is remote-first, so it can be executed from anywhere in Europe, with the ability to operate in GMT and GMT+1 timezones. We're building agents that run real parts of our business, take on real work, and get better based on the results they produce.
Evidence noteDescribe an agent or LLM-powered system you have shipped. How did you evaluate whether it was working?
Wanted
Mistral AI
Software Engineer (Golang), Sandbox
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Location: Paris. Location Type: Remote. The listing gives no further remote-eligibility or attendance text.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[3]
Location is unconfirmed: the role is labelled Remote with a Paris location, but the listing does not state which countries remote applicants may work from.[3]
AI centrality is not established: the role builds secure, isolated sandboxes (microVMs or containers, snapshot and restore) designed for AI agents and reinforcement-learning workloads. The primary duties are Linux isolation, security boundaries and fleet-scale infrastructure in Go rather than building or evaluating AI models.[3]
No level is named; the listing asks for 5+ years building backend or infrastructure systems and production Go experience. The application form rendered on the check date with Mistral’s 3-applications-in-90-days limit.[3]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: part of the Cloud Platform team building Mistral’s European AI cloud. The Sandbox team builds network-controlled execution environments where customers’ agents and RL training runs execute untrusted code, created and destroyed at high rates with low latency. Separate product diligence was not advanced because the role is not ranked.[3]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteWe build secure, isolated, network-controlled environments with snapshotting primitives, designed for AI agents and reinforcement-learning workloads. Candidates may submit up to 3 total applications within a 90 day period across all open roles.
Wanted
Mistral AI
Research Engineer, Orchestral
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Location: Paris; London; Warsaw. No hybrid, remote or office-attendance policy is stated on the listing.[4]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[4]
Location gate unverified: Paris is named, but no attendance policy is given.[4]
AI centrality passes, with a caveat: the role builds the agentic platform Mistral’s researchers use to train and evaluate models, including the rollout engine, long-running tasks, subagents and RL features. The listing itself calls it a software-engineering role, with RL and PyTorch experience not required.[4]
No level is named; the listing asks for experience building and debugging complex distributed systems at scale and strong Python. First published October 16, 2026, it is new to this site’s history, not newly created. The application form rendered on the check date.[4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Orchestral is the internal system Mistral’s science team uses to run training and evaluation at scale, from the rollout engine behind agentic workflows to the infrastructure that keeps those workloads reliable. Separate product diligence was not advanced because the role is not ranked.[4]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteAs a Research Engineer on the Orchestral team, you will help build the agentic platform we use to train and evaluate our models. Candidates may submit up to 3 total applications within a 90 day period across all open roles.
Pay note“Estimated Base Salary Band €65K – €170K”. The listing says most roles are paired with “meaningful equity” but does not state equity for this role.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralfail
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published base-salary band starts at €65,000, below both the €100,000 floor and the €80,000-plus-equity branch, and equity is not stated for this specific role.[5]
AI-central engineering gate fails: the role is for tax or audit professionals who demonstrate the product, run pilots and build client workflows with Legora’s AI workspace, not for building or operating AI systems.[5]
Location fails the hybrid rule as written: the role is on-site in Paris rather than hybrid. The application form rendered on the check date.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Legora sells an AI workspace for legal professionals; this Paris role brings it to tax and audit teams through demonstrations, pilots, onboarding and reusable workflow templates.[5]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteEstimated Base Salary Band €65K – €170K. As a Legal Engineer in our Paris team, you’ll help tax and audit professionals put AI to work in their day-to-day practice.
Wanted
OpenAI
Applied AI Engineer, Digital Natives
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris, France. Location Type: Hybrid. “This role is based in Paris. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees”[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing, and the official ATS data carries no compensation field for this role. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual cash-base lower bound was found, so the role cannot be ranked under the pay rule.[1]
Location passes on employer evidence: Paris with three office days per week.[1]
AI centrality passes: the engineer designs, builds and deploys AI systems with customers, working hands-on in code on prototypes, evaluation harnesses, integrations and production accelerators across models, agents, retrieval and tools.[1]
No level is named in the title or requirements; the listing asks for a demonstrated track record of taking AI or ML systems from prototype to production, and fluency in French and English. The role sits in the Go To Market department.[1]
The application form rendered on the check date and states a limit of five applications in any 180-day span.[2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: OpenAI’s Applied AI Engineering team helps enterprise and digital-native customers turn frontier AI capabilities into production systems, from use-case selection and architecture through prototyping, evaluation, launch and scale, and feeds deployment lessons back to OpenAI product teams. Separate product diligence was not advanced because the role is not ranked.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is based in Paris. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees. Work hands-on in code to build prototypes, evaluation harnesses, reference implementations, integrations, and production accelerators. Are fluent in both French and English
Evidence notePlease Note: we have set up limits for applications across roles. Candidates may not apply more than 5 times in any 180 day span.
Wanted
Mistral AI
Research Engineer, Latency & Throughput
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris; London. “This role is based in our Paris or London offices (hybrid).” “For senior profiles, we will also consider remote candidates based in France.” Remote hires visit the Paris office for the first onboarding week, “Then at least 3 days per month”. No weekly office attendance is stated for the hybrid option.[3]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[3]
Location is conditional: hybrid attendance is not quantified, and the remote-from-France option applies only to “senior profiles”, a level the listing does not define.[3]
AI centrality passes: the team researches and prototypes LLM inference techniques such as speculative decoding, quantization, KV compression and new attention variants, then ships those that prove out into the serving stack.[3]
The role is research-leaning: the listing asks for research depth such as a thesis, publications or efficiency-oriented lab work, plus hands-on Python and PyTorch.[3]
The application form rendered on the check date; it limits candidates to 3 applications in 90 days across all open roles.[3]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: the Latency & Throughput team sits between Mistral’s production inference engineers and its research scientists, turning recent inference-efficiency literature into experiments and shipped serving-stack improvements. Separate product diligence was not advanced because the role is not ranked.[3]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is based in our Paris or London offices (hybrid). For senior profiles, we will also consider remote candidates based in France. Candidates may submit up to 3 total applications within a 90 day period across all open roles.
Wanted
Mistral AI
Applied Scientist - Research Engineer (Formal Methods), EMEA
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Location: Paris; London. No hybrid, remote or office-attendance policy is stated on the listing.[4]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[4]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published salary range.[4]
Location gate unverified: Paris is named, but no attendance policy is given.[4]
AI centrality passes: run pre-training and post-training of large language models, curate theorem-proving and software-verification datasets, and build evaluators and benchmarks for formal reasoning. No seniority level is stated.[4]
The application form rendered on the check date with the same 3-applications-in-90-days limit.[4]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: applied science on formal methods at Mistral AI, training and deploying LLMs for theorem proving and software verification and managing research projects with client research teams. Separate product diligence was not advanced because the role is not ranked.[4]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteGenerate and curate datasets for theorem proving and software verification, and build evaluators and benchmarks for formal reasoning. Candidates may submit up to 3 total applications within a 90 day period across all open roles.
Wanted
Hugging Face
Senior Open-Source Python Engineer, ML Developer Tools - EMEA Remote
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Title: “EMEA Remote”. The rendered page shows “Remote” without naming eligible countries; France appears only as a hidden location in the official Workable data. “While we have office spaces in NYC and Paris, we're very distributed”.[5]
Salary evidence
No published figures
Equity
Yes
Pay note“All employees have company equity as part of their compensation package.” No salary range is published; pay is unknown, not below the threshold.[5]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: equity is stated, but no cash-base range is published.[5]
Location is unconfirmed: “EMEA Remote” is stated, but the visible listing does not name France or country eligibility.[5]
AI centrality is not established: the role maintains developer libraries for building ML applications (Gradio, Trackio, FastRTC), including real-time streaming and frontend work. It builds tools for ML developers rather than building, evaluating or operating AI systems as the primary duty.[5]
Seniority is explicit in the title, with 5+ years of professional Python required. First published September 21, 2026, it is new to this site’s history, not newly created. The application form rendered on the check date.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: the Applications team builds tooling for creating ML applications on top of Hugging Face Hub primitives, maintaining Gradio, the Trackio experiment tracker and the FastRTC streaming library, with real-time streaming named as a priority for Gradio 7. Separate product diligence was not advanced because the role is not ranked.[5]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence notemaintaining and growing libraries like Trackio, a local-first, agent-friendly experiment tracking platform, FastRTC, our library for real-time audio and video streaming in Python, and Gradio itself. All employees have company equity as part of their compensation package.
Wanted
Alan
Security Engineer Senior - GRC
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Anywhere in France. Location Type: Hybrid. No office-attendance frequency is stated.[6]
Salary evidence
Known annual EUR base
Equity
Yes
Pay noteCompensation “Level D to E”: €83K – €100K, classified as an annual EUR salary in the official ATS data. “At Alan we offer an attractive Equity package on top of an above market-average base salary.”[6]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralfail
Paypass
Product researchunknown
Gate notes
AI-central gate fails: the role owns security governance, risk and certification (ISO 27001, HDS) and incident governance. AI appears only as one risk area to govern.[6]
Pay would pass the €80,000-plus-equity branch: an €83,000 annual floor with an equity package stated on the listing.[6]
Hybrid “Anywhere in France” is stated, but attendance frequency is not. The application form rendered on the check date.[6]
Evidence note€83K – €100K. At Alan we offer an attractive Equity package on top of an above market-average base salary. You map and prioritise security risks, including AI risks, turn them into clear treatment plans
Wanted
Langfuse (ClickHouse)
Langfuse - Senior Product Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Location: Germany; France; Italy; Portugal; Spain; The Netherlands. ATS location type: Hybrid. Langfuse careers page: “European roles are remote-first with 3-5 days every two months in Berlin (travel covered); SF and Berlin locals use the office when it's useful.”[8][3]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“Salary €90K – €160K • Offers Equity”, classified as an annual EUR salary in the official ATS data; every new team member receives stock options.[8]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Paypass
Product researchunknown
Gate notes
AI centrality is not established: the role is full-stack feature ownership across a React frontend, Next.js backend, ClickHouse analytics, SDKs and docs on an LLM engineering platform. Building AI systems is not stated as the primary duty, and ML/AI background is listed as a bonus.[8]
Pay and location would pass on the same evidence as the SDK role: a €90,000 annual floor with equity, and remote-first work from France with periodic Berlin visits.[8][3]
It was first published on September 18, 2026 and is new to this site’s history. The same-page application form rendered on the check date.[8]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: end-to-end product features for Langfuse, from user calls and specs to code, docs and changelog. Named past examples include agent trace views and agent graphs, the TypeScript SDK v4 rewrite, and dashboard management via API, CLI and MCP.[8]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteWork across the full stack: and we mean full. ML/AI background or familiarity with the LLM framework ecosystem
Wanted
Mistral AI
Evaluation Engineer, Applied AI
Near-miss
Report date
Evidence checked
Posted
Work location
Unknown — Location: Paris. No hybrid, remote or office-attendance policy is stated on the listing.[10]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published, and the official ATS data carries no compensation. Pay is unknown, not below the threshold.[10]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay is unpublished, so the pay gate cannot be verified.[10]
Paris is named, but no attendance policy is given.[10]
AI centrality passes: the role designs frameworks to evaluate LLM performance across customer use cases and builds evaluation infrastructure. The listing asks for at least 3 years of evaluating ML, LLM or agentic systems and does not name a senior level.[10]
The application form rendered on the check date; it limits candidates to 3 applications in 90 days across all open roles.[10]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: on the Applied AI team, build customer-specific evaluation suites and scalable evaluation pipelines that define when an LLM is ready for production, and feed results back to research and product teams.[10]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign and implement comprehensive frameworks to evaluate large language model (LLM) performance across customer use cases. At least 3 years of experience evaluating ML, LLM, or agentic systems. Candidates may submit up to 3 total applications within a 90 day period across all open roles.
Wanted
White Circle
AI Red Team Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Location: Global. Location Type: Remote. France is not named, and no country eligibility rules are stated.[11]
Salary evidence
No published figures
Equity
Yes
Pay note“Competitive compensation package, including equity”. No salary range is published; pay is unknown, not below the threshold.[11]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay is unpublished; equity is mentioned without a cash-base range.[11]
Location is unconfirmed: the listing says “Global” remote but does not explicitly name France or state country eligibility.[11]
AI centrality passes: hands-on adversarial testing of LLM systems, including prompt injection, tool misuse and data leakage, with Python attack automation and regression tests. No seniority level is stated.[11]
The application form rendered on the check date.[11]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: White Circle builds a safety, reliability and optimization layer for AI systems based on natural-language policies. This role red-teams chatbots, copilots, RAG pipelines and agents and turns findings into an internal attack library, reports and regression tests.[11]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteRed-team LLM-powered systems: chatbots, copilots, RAG pipelines, AI agents, tool-calling workflows, and API-based AI products. Competitive compensation package, including equity
Wanted
Langfuse (ClickHouse)
Langfuse - Senior Backend Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Location: Germany; France; Italy; Portugal; Spain; The Netherlands. ATS location type: Hybrid. Langfuse careers page: “European roles are remote-first with 3-5 days every two months in Berlin (travel covered); SF and Berlin locals use the office when it's useful.”[9][3]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“Salary €90K – €160K • Offers Equity”, classified as an annual EUR salary in the official ATS data.[9]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Paypass
Product researchunknown
Gate notes
AI-central gate fails: the primary duties are the trace-ingestion pipeline, ClickHouse schemas and query performance. LLM-as-a-Judge appears as one example of infrastructure-heavy features, not the core assignment.[9]
Pay and location would pass on the same evidence as the SDK role.[9][3]
The same-page application form rendered on the check date.[9]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: own the ingestion pipeline that moves terabytes of AI tracing data per day from the API into ClickHouse, with load tests on production data.[9]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteOwn and optimize our ingestion pipeline. Build new infrastructure-heavy features: things like LLM-as-a-Judge
Wanted
Anthropic
Manager of Applied AI Architecture, Industries
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France. “Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.”[12]
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay noteAnnual Salary: €265.000 - €310.000 EUR. The listing adds: “For sales roles, the range provided is the role’s On Target Earnings ("OTE") range”. Whether this figure is base or OTE is not confirmed.[12]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Hands-on AI engineering gate fails: this is a second-level manager role that hires and manages first-line managers of pre-sales Applied AI Architects across France, Iberia and Italy.[12]
Pay is not confirmed as cash base, because the listing says sales-role ranges are OTE.[12]
The same-page application form rendered on the check date.[12]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: lead the Southern Europe Applied AI Industries pre-sales architecture organization and sponsor strategic enterprise customers adopting Claude products.[12]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteHire, manage, and guide a team of 1st line managers who will in turn build teams of pre-sales Applied AI Architects across France, Iberia and Italy. For sales roles, the range provided is the role’s On Target Earnings ("OTE") range. €265.000 - €310.000 EUR. Currently, we expect all staff to be in one of our offices at least 25% of the time.
Pay noteCompensation tier “Lead SE- Full Comp - FR”: €130K – €160K. The “Full Comp” label leaves base versus total package unconfirmed.[13]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Hands-on AI engineering gate fails: this is a founding player-coach role that leads the global Solutions Engineering function across pre-sales and post-sales, with authority over hiring, structure and operating model.[13]
The location gate fails because the role is on-site rather than hybrid or remote. Pay is not confirmed as cash base.[13]
First seen in this site’s ATS pay sweep today; the requisition was first published on August 3, 2026.[13]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: build and lead Dust’s Solutions Engineering organization for enterprise evaluations and post-deployment expansion of its AI agent platform.[13]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis is a founding player-coach role. You will lead our global Solutions Engineering function across pre-sales and post-sales. Lead SE- Full Comp - FR
Wanted
OpenAI
Forward Deployed Engineer - Paris
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris, France. Location Type: Hybrid. “This role is based in Paris. We use a hybrid work model of 3 days in the office per week. We offer relocation assistance. Travel up to 50% is required.”[1]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing, and the official ATS data carries no compensation field for this role. Pay is unknown, not below the threshold.[1]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: no employer-published annual cash-base lower bound was found, so the role cannot be ranked under the pay rule.[1]
Location passes on employer evidence: Paris with three office days per week, plus up to 50% travel. The application asks: “Are you able to work from our Paris office three days per week?”[1][2]
AI centrality passes: FDEs “lead complex end-to-end deployments of frontier models in production” and are asked to have built or deployed systems powered by LLMs or generative models.[1]
Level is not named in the title; the listing asks for 5+ years of engineering or technical deployment experience that includes customer-facing work.[1]
The application form rendered on the check date and states a limit of five applications per 180 days across roles.[2]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: OpenAI’s Forward Deployed Engineering team partners with strategic customers to put frontier models into production. The role owns discovery, technical scoping, system design, build and production rollout, measured by production adoption and eval-driven feedback to product and model roadmaps. Separate product diligence was not advanced because the role is not ranked.[1]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is based in Paris. We use a hybrid work model of 3 days in the office per week. We offer relocation assistance. Travel up to 50% is required. Forward Deployed Engineers (FDEs) lead complex end-to-end deployments of frontier models in production alongside our most strategic customers.
Remote — Remote Ireland; Remote, France; Remote, Germany; Remote, Netherlands; Remote, United Kingdom. “All of our roles are remote, however some roles may carry specific location-based eligibility requirements.”[3]
Salary evidence
No published figures
Equity
Yes
Pay noteNo base-salary range is published for France. The benefits section lists “Equity Compensation & Employee Stock Purchase Plan”; equity is not counted as cash. Pay is unknown, not below the threshold.[3]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Pay gate unverified: the official listing and Greenhouse pay-transparency data publish no base range for this role.[3]
Location passes on employer evidence: remote with France explicitly named among the eligible countries.[3]
Seniority is explicit in the body title: Staff Forward Deployed Engineer, Agentic SDLC (EMEA only).[3]
AI centrality is partial: driving production adoption of GitLab Duo Agent Platform is one of three duty areas, alongside production code contributions to GitLab’s product and strategic customer stewardship. Strong Ruby on Rails or Go experience is requested.[3]
Greenhouse first-published date is May 11, 2026; the posting was updated on October 15, 2026. The same-page application form rendered on the check date.[3]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: a field-situated product engineering role for strategic GitLab customers. It contributes production-quality fixes and features to GitLab, diagnoses complex adoption and scale problems, and develops customer-safe patterns for the agentic software development lifecycle, AI governance and human-in-the-loop delivery with GitLab Duo Agent Platform. Separate product diligence was not advanced.[3]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteStaff Forward Deployed Engineer, Agentic SDLC - (EMEA Only). Drive governed, measurable, and production-oriented adoption of GitLab Duo Agent Platform with strategic customers. Strong Ruby on Rails or Go experience and the ability to contribute directly to GitLab’s product and adjacent systems.
Wanted
OpenAI
Applied AI Engineer, Codex
Rejected
Report date
Evidence checked
Posted
Work location
Unknown — ATS location: Paris, France, Hybrid. The description says: “This role is based in our London office. We use a hybrid work model of 3 days in the office per week.” The application asks about working “from a local office three days per week”.[4][5]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing or in the official ATS data. Pay is unknown, not below the threshold.[4]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Location gate fails as ambiguous: the ATS location is Paris, but the description says the role is based in the London office. The Paris attendance policy cannot be confirmed from employer evidence.[4]
Pay is unpublished.[4]
Hands-on AI work is central: building AI coding workflows, integrations and evaluation harnesses for Codex deployments. Fluent English and French are requested. This is a different requisition from the Codex startups role recorded on October 14.[4]
The application form rendered on the check date.[5]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: partner with engineering organizations to design, build and deploy AI-powered software development workflows with Codex, including evaluation of AI coding systems with representative tasks and automated graders.[4]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role is based in our London office. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees. Design systematic approaches for evaluating AI coding systems using representative software engineering tasks, automated graders, production signals, and developer feedback.
Evidence noteAre you able to work from a local office three days per week?
Wanted
Databricks
Sr. Forward Deployed Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Unknown — Paris, France. No hybrid, remote or office-attendance policy is stated; the listing says “Travel to customers 20% of the time”.[6]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo salary range or equity terms are published on the listing, and Greenhouse pay-transparency data is empty. Pay is unknown, not below the threshold.[6]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Location gate fails as ambiguous: Paris is named, but no attendance policy is published.[6]
Pay is unpublished.[6]
AI centrality is not established: the requirements centre on data engineering, Apache Spark internals and cloud platforms, with “working knowledge of MLOps, ML/AI models and AI APIs”. The embedded application form rendered on the check date.[6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: billable, customer-facing delivery of production data and AI systems on the Databricks platform, spanning data engineering, ML/AI model integration and applications.[6]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence note6+ years experience in data engineering, data platforms & analytics, or software engineering. Deep experience with distributed computing with Apache Spark™ and knowledge of Spark runtime internals. Working knowledge of MLOps, ML/AI models and AI APIs. Travel to customers 20% of the time
Wanted
Celonis
Senior/ Lead AI Deployment Engineer - Supply Chain
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France. The listing mentions “a flexible hybrid work model” and client-site travel “primarily based in Paris, typically 2–3 days per month”; no weekly office-day count is published.[7]
Salary evidence
No published figures
Equity
Yes
Pay noteNo base-salary range is published; the application asks for a salary expectation. Every full-time employee receives Restricted Stock Units (RSUs) according to the listing. Pay is unknown, not below the threshold.[7]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-central gate fails: the role delivers process-mining solutions on the Celonis platform (data models, process analyses, dashboards and business cases), using AI tools to accelerate the build rather than building AI systems as the primary duty.[7]
Pay is unpublished, and the weekly office attendance is not stated.[7]
The same-page application form rendered on the check date. Professional French and English and supply-chain domain experience are required.[7]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: hands-on customer delivery of supply-chain process-intelligence solutions on the Celonis platform, directing Celonis’s AI deployment tooling and validating its outputs.[7]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteBuild and configure the analytical core on the platform—data models, process analyses, and dashboards —directing Celonis's AI deployment tooling to move fast and validating every output for accuracy and business relevance.
Wanted
Algolia
Senior Site Reliability Engineer, AI Platform
Rejected
Report date
Evidence checked
Posted
Work location
Unknown — Paris, France. “Positions listed within a specific city are only available in that location - depending on the role it may be available with either a hybrid-remote or in-office schedule.” No schedule is given for this role.[8]
Salary evidence
Known annual EUR base
Equity
Unknown
Pay noteBase Salary Pay Range €69.768 - €96.900 EUR. No equity terms are stated in the listing.[8]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralunknown
Payfail
Product researchunknown
Gate notes
Pay gate fails: the published annual base floor of €69,768 is below €80,000, so it fails both the €100,000 branch and the €80,000-with-equity branch.[8]
The office schedule is not stated for this Paris role.[8]
The role operates Kubernetes and cloud infrastructure for AI-related workloads; AI/ML serving experience is listed only as nice to have. The same-page application form rendered on the check date.[8]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: Algolia’s AI Platform team builds and operates shared production foundations for AI-powered capabilities, covering Kubernetes, cloud infrastructure, CI/CD, observability, reliability and FinOps.[8]
Source evidence notes (1)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteOwn and evolve production infrastructure supporting AI-related workloads and services at scale. Base Salary Pay Range €69.768 - €96.900 EUR
Wanted
Sierra
Software Engineer, Agent
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Paris; ATS location type On-site. The application asks: “We are primarily an in-person company (Monday-Thursday). Does that work for you?”[7]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note€125K – €265K • Offers Equity. Official ATS data lists a one-year Salary component plus an equity component.[6]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Paypass
Product researchunknown
Gate notes
The level is not named. The listing title and requirements do not state senior or staff, so the seniority gate is unverified even though the salary floor exceeds the threshold.[6]
Paris attendance is explicit: the application asks for Monday–Thursday in-person work. The role has been open since December 2025.[7]
Building and evolving production customer-facing AI agents is central to the role. Separate company and product diligence was not advanced because the role is not ranked.[6]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: design and deliver production-grade AI agents for enterprise customers in finance, healthcare and commerce, and own the Agent Development Life Cycle from pilot to iteration. Separate product diligence was not advanced.[6]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteWe are primarily an in-person company (Monday-Thursday). Does that work for you?
Wanted
Doctolib
Senior Agents Platform Engineer (x/f/m)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Paris; Hybrid. “Hybrid work setup (up to 2 remote days per week)”.[8]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical salary or equity terms are published on the listing.[8]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
The annual gross cash-base floor is unknown, not below threshold: no pay band is published.[8]
AI-central senior platform work: an agent manifest and registry, a LiteLLM-based GenAI gateway, long-running agent infrastructure, guardrails, and the evaluation and audit layer for healthcare agents.[8]
The application form rendered. The ATS employment-type label reads Contract while the listing describes French employee benefits; confirm the contract type.[13]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: a shared agent runtime, a unified GenAI gateway serving several hundred million requests per month, and observability and offline evaluation (LLM-as-judge, patient and practitioner simulators) for Consultation Assistant, Phone Assistant and Health Companion. Separate product diligence was not advanced.[8]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence notePlease tell us if you have been a Public Official or a Close Family Member of a Public Official in the last 5 years?
Wanted
Doctolib
Senior IA Inference Engineer (x/f/m)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Paris; Hybrid. “Hybrid work setup (up to 2 remote days per week)”.[9]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical salary or equity terms are published on the listing.[9]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
The annual gross cash-base floor is unknown, not below threshold: no pay band is published.[9]
AI-central senior work: own the self-hosted LLM serving platform for open-weight models, and tune vLLM, Triton and TensorRT-LLM for latency and cost across an L4-to-H200 GPU fleet.[9]
The application form rendered. The ATS employment-type label reads Contract; confirm the contract type.[14]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: shared LLM serving foundations on Ray Serve / Anyscale, a model registry and promotion path, inference optimisation, SLOs and cost reporting that keep sensitive patient data on infrastructure Doctolib controls. Separate product diligence was not advanced.[9]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence notePlease tell us if you have been a Public Official or a Close Family Member of a Public Official in the last 5 years?
Wanted
Pennylane
Senior Machine Learning Engineer - Embedded AI
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — All France (remote); Germany (remote); Italy; Portugal (remote); Spain; Spain (remote). Remote location type.[11]
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical salary or equity terms are published; the application asks for salary expectations.[11]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
The annual gross cash-base floor is unknown, not below threshold.[11]
Explicit France-remote eligibility. Senior ML engineering for document understanding, extraction, matching and scoring behind Copilot and Autopilot; 5–8 years of experience requested.[10]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: specialised ML systems behind Copilot and Autopilot (invoice parsing, document classification, accounting suggestions, matching and scoring), owned across the full lifecycle. Separate product diligence was not advanced.[10]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteWe build the specialized machine learning systems behind Copilot and Autopilot. Have 5–8 years of experience and are very strong in Python.
Evidence noteAll France (remote). What are your salary expectations ?
Wanted
Escape
Senior Full Stack Engineer | AI & Security
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Paris Office; Hybrid. “Location: Paris (75002), 2 days remote per week”.[12]
Salary evidence
Known annual EUR base
Equity
Yes
Pay note“Compensation: €75–90k base + 0.05% BSPCE depending on seniority and scope”.[12]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Payfail
Product researchunknown
Gate notes
Pay fails: the published base floor is €75,000. Equity is explicit, but the floor is below the €80,000 equity-path minimum.[12]
AI-central assignment is unverified: the employer describes PostgreSQL performance, data modelling and full-stack product work as most of the job, with AI-native workflow surfaces as one part. This is a different requisition from the previously reported Escape research role.[12]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteListing summary: product and platform engineering for an AI offensive-security product. Owns the asset and findings data model, Svelte and GraphQL surfaces, and AI-assisted triage and remediation workflows. Separate product diligence was not advanced.[12]
Source evidence notes (2)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteCompensation: €75–90k base + 0.05% BSPCE depending on seniority and scope. Push PostgreSQL to its edge. Location: Paris (75002), 2 days remote per week.
Remote — Official ATS metadata: remote true; workplace remote; Paris, France; countryCode FR. France eligibility is explicit rather than inferred from EMEA.
Salary evidence
No published figures
Equity
Yes
Pay noteNo numerical cash-base floor is published. All employees have company equity as part of their compensation package.[20]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Annual gross cash-base lower bound is unknown, not published below threshold. Equity alone cannot satisfy the cash requirement.[20]
Senior ownership of open voice-agent infrastructure and production multimodal systems is explicit; the rendered form asks about Voice/ASR/TTS experience.[22][21]
Evidence noteDo you have hands-on experience in Voice/ASR/TTS ?
Wanted
Dust
Senior AI Support Engineer (Paris)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Paris. Role header: On-site. Office-first description allows occasional home-working; application asks: Are you happy 4 days office based?
Salary evidence
No published figures
Equity
Yes
Pay noteCompetitive compensation; Significant equity package in a Sequoia-backed startup. No numerical cash-base floor published.[26]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
No numerical annual gross cash-base floor is established; competitive compensation and equity do not pass the pay gate.[26]
A senior individual contributor role that builds support agents, retrieval, escalation and measurement systems; it also includes front-line complex support work.[26]
Four office days are explicitly asked in the application. Occasional home-working in the description does not imply a more flexible schedule.[27]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: build ticket-classification, drafting, routing, incident-detection and knowledge-retrieval agents; debug model responses and close feedback loops. Named workstream is support automation, not frontier-model training. Separate product diligence not advanced.[26]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Remote — Remote-first Europe; role location list includes Berlin and other named European countries but does not name France. France eligibility remains unestablished.
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay noteWestern Europe / Northern Europe: €114.3K – €149.3K • Offers Equity. ATS Salary interval: 1 YEAR. Regional tier shown for transparency, not a verified French offer; gross cash-base classification for France remains unconfirmed.[28][13]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
France is absent from the role-specific location list and structured secondary locations. Europe-wide branding and a general country selector do not establish France eligibility.[28][13]
Do not transfer the Western/Northern Europe salary tier to a French offer. The application defines expected compensation as gross yearly base, but this is an applicant-expectation prompt, not a French salary commitment.[29]
AI-primary platform duties are explicit. Senior/staff grade is not named; first-of-kind technical leadership is expected, but exact level remains unknown.[28]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: isolated agent execution, permissions/secrets guardrails, issue-to-draft-PR workflows, repository playbooks, CI integration and benchmarks for code-agent quality, failures and reviewer burden. Separate product diligence not advanced.[28]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDesign and implement isolated execution environments where agents can work without exposing secrets or accessing sensitive systems unnecessarily.
Evidence noteWe're assuming this answer to be the gross yearly base salary for a full time permanent contract. If your role includes a variable, please detail.
Evidence noteMetadata summary: published September 24, 2026; Remote; no France secondary location; Western/Northern Europe Salary 114300–149300 EUR, interval 1 YEAR.
Wanted
Qonto
Senior Android Engineer - AI Product
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — France is explicitly among permitted countries. The application offers a France-based 100% remote preference, qualified by: we will discuss it during the hiring process. Final arrangement is not established.
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical annual gross cash-base floor or explicit equity grant found in the examined role/application.[30]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Pay is unpublished, not demonstrated insufficient.[30]
AI-primary system-building remains unresolved: remit emphasizes Kotlin Android user experiences for AI agents, while the listing describes AI engineering as a future growth path.[30]
The form explicitly includes France-based remote applicants, but asks an ideal arrangement to discuss during hiring; unconditional full-remote terms are not verified.[31]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: Android UI/product delivery for finance-agent experiences, reliable Kotlin architecture and mobile parity with web. No individually assigned model-training or agent-evaluation system is established. Separate product diligence not advanced.[30]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteShip features, fast & well: Deliver Android features for our AI agents from scoping to release, balancing speed with reliability and minimal technical debt.
Evidence noteWhat working arrangement are you ideally looking for? (we will discuss it during the hiring process)
Wanted
Thunderbird
Staff Product Engineer, Enterprise AI
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Remote; employer publishes only Canada and US pay markets and requires residence and permanent work authorization in the specified markets. France is not established; European customer hours are not hiring eligibility.
Salary evidence
No published figures
Equity
Unknown
Pay noteCanada: $155,000 - $185,000 CAD; US: $185,000 - $220,000 USD. No France salary published. Foreign market ranges are not stored as French pay or converted.[32]
Recorded gates (as checked on the report date)
Listing openpass
Locationfail
AI-centralpass
Payunknown
Product researchunknown
Gate notes
France-based eligibility is not established. The employer explicitly cannot consider applicants outside its specified markets; regular European-hour overlap is not authorization to work from France.[32]
Staff hands-on production engineering is explicit, including LLM platforms, RAG, agents and enterprise deployment. Actual employer is Thunderbird Technologies Corporation, separate from Mozilla Corporation/Foundation.[32]
Pay for a French contract is unknown, not a below-threshold offer. No currency conversion is appropriate for a geographically inapplicable band.[32]
Evidence noteWhat interests you about this role and the opportunity to work on Thunderbolt at Thunderbird?
Wanted
Cohere
Member of Technical Staff, Training Infra Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Paris is an explicit role location. This role can be based remotely or from one of our office locations listed on the job description - there is no minimum in-office qualification requirement.
Salary evidence
No published figures
Equity
Unknown
Pay noteNumerical annual base tiers are published only for Canada and the USA, with equity. No French numerical base floor or France equity applicability is established; regional tiers are not transferred.[12][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
France remote eligibility is explicit through Paris plus no minimum office attendance. The application form renders.[12][45]
France-specific annual gross cash-base floor is unpublished; CAD/USD North American tiers cannot pass a French pay gate.[12][8]
Hands-on training infrastructure and large-model distributed training are central. Exact senior/staff grade is not named and remains unresolved.[12]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: high-performance training software, training-cycle tooling, supercompute/data infrastructure and production model delivery. Python, JAX/PyTorch, XLA/MLIR, Kubernetes/Slurm and Ray are named; large-scale training experience is required. Named internal project unknown; separate product diligence not advanced.[12]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteThis role can be based remotely or from one of our office locations listed on the job description - there is no minimum in-office qualification requirement.
Evidence noteAre you authorized to work in the country you currently reside in?
Wanted
Anthropic
Forward Deployed Engineer
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France. Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical annual gross cash-base range is published in the role page. Generic competitive compensation and optional equity donation matching do not establish a qualifying base floor or an explicit equity grant.[42]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Numerical annual gross cash-base lower bound is missing; this is unknown compensation, not a published below-threshold salary.[42]
Hands-on production Claude applications, MCP servers, sub-agents and agent skills are central duties. The listing requests 8+ years in technical customer-facing work and production LLM experience, with 25–50% customer-site travel.[42]
Paris and a minimum 25% office policy are explicit, but the board's On-Site label differs from the page's hybrid heading; confirm the actual attendance schedule. The application renders and asks about office attendance and prior enterprise LLM deployments.[42][41]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: build and deploy customer production applications on Claude, ship MCP servers, sub-agents and agent skills, support enterprise deployments and codify repeatable deployment patterns. The specific customer project is unknown. This is hands-on model integration and agent deployment, not a frontier-model-training role. Separate product diligence was not advanced.[42]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteLocation-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.
Evidence noteMetadata summary: requisition 5391021008; first published August 17, 2026; Paris, France; separate Location Type metadata On-Site. The role-page hybrid policy is recorded alongside this discrepancy.
Wanted
Anthropic
Applied AI Engineer, Enterprise
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France. Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay noteAnnual Salary: €220.000 - €235.000 EUR. The same page says the range for sales roles is OTE, including commissions/bonuses and annual base. This pre-sales/advisory role does not provide a separate cash-base floor; the displayed range is not confirmed base.[37]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Requested hands-on AI-engineering remit is not established: the role owns technical advisory, proof of value, executive business cases and procurement, while a systems integration partner does the build. AI subject matter alone does not establish the required hands-on assignment.[37]
The €220,000–235,000 annual compensation range cannot be treated as cash base in light of the sales OTE disclaimer and absent base breakdown. Pay remains unknown, not below threshold.[37]
The page requests 6+ years of technical experience, pre-sales/value-based selling experience and native or fluent French. Paris minimum office attendance is explicit; the same-page application form is reachable.[37][41]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: advise large French enterprises on Claude architecture, agent design, evaluation, security and adoption. Duties emphasize technical wins, executive communication, proof of value and systems-integrator enablement rather than personally owning production implementation. Named customer project unknown; separate product diligence was not advanced.[37]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteOnce they have decided, you set their systems integration partner up to deliver, agreeing the target architecture and what good looks like, and you stay close as the customer's technical advisor on Claude throughout deployment, while the partner does the build.
Evidence noteFor sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role.
Evidence noteDo you have experience in a pre-sales or technical advisory role selling into large enterprises, ideally in the French market, including scoping and designing complex solutions?
Evidence noteMetadata summary: requisition 5390754008; first published August 20, 2026; Paris, France; Location Type On-Site. Exact weekly attendance beyond the page's minimum is not established.
Wanted
Anthropic
Applied AI Architect, Partnerships
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France. Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Unknown
Pay noteAnnual Salary: €190.000 - €220.000 EUR. The same page identifies sales-role ranges as OTE, including commissions/bonuses and base. No separate annual gross cash-base floor is established.[43]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
The employer explicitly describes a pre-sales architect role for systems integrators and cloud partners, oriented toward indirect revenue, partner enablement, customer-deal support and events. This is not the requested senior/staff hands-on AI-engineering assignment.[43]
Annual compensation is published but is not verified cash base because the sales OTE disclaimer applies to the stated pre-sales context and no base breakdown is supplied.[43]
Five-plus years in technical customer/partner-facing roles are requested. The application is reachable; Paris and minimum 25% office attendance are explicit, with separate ATS On-Site metadata retained as a caveat.[43][41]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: Claude partner solution architecture, GenAI reference designs and cloud/system-integrator enablement involving AWS and GCP. A primarily owned model, evaluation system or production-agent project is not specified. Separate product diligence was not advanced.[43]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteAs an Applied AI Architect on the Partnerships team at Anthropic, you will be a pre-sales architect focused on cultivating technical relationships with our Global and Regional System Integrators (GSIs/RSIs), and our cloud partners (AWS and GCP).
Evidence noteFor sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role.
Evidence noteMetadata summary: requisition 5390787008; first published August 17, 2026; Paris, France; separate Location Type metadata On-Site.
Wanted
DualEntry
Applied Engineer (Remote)
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Remote (EU, LATAM, Canada). Remote-first team. The application requires working hours overlapping the Eastern Time Zone (NYC); France is not explicitly named.
Salary evidence
Known annual base, non-EUR (no conversion applied)
Equity
Yes
Pay noteBase salary: $150,000 - $280,000; Equity: $50,000 - $80,000. Employer ATS identifies USD Salary with interval 1 YEAR. Original currency retained; no EUR comparison or positive pay-gate decision is made after the scope/location exclusions. Equity is not counted as cash.[44][9]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralfail
Payunknown
Product researchunknown
Gate notes
Primary duties are Temporal migration/integration pipelines, Postgres accounting-data reconciliation and internal tooling. AI-native ERP branding and access to AI tools do not establish AI-primary engineering responsibilities for this hire.[44]
Remote EU/LATAM/Canada wording does not explicitly establish France-based applicant/payroll eligibility. The application adds Eastern Time working-hours overlap.[44][45]
Existing requisition identity f3d95a4c-dc28-4761-a8bc-66ab870d7121 is retained. The October 15 publication timestamp does not by itself prove a new role; this record is distinct from the archived Senior AI Engineer requisition and is not a replacement qualified selection.[9]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: finance ERP customer delivery, including idempotent/resumable/auditable Temporal workflows, accounting data migration into Postgres, reconciliation and productized internal tooling. Five-plus years shipping full-stack production software are requested. No model-training, agent-development or evaluation assignment is documented; separate product diligence was not advanced.[44]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteCurrent metadata summary: requisition f3d95a4c-dc28-4761-a8bc-66ab870d7121; published October 15, 2026; Applied Engineer (Remote); Remote (EU, LATAM, Canada); USD Salary 150,000–280,000, interval 1 YEAR; equity is a separate component. A refreshed timestamp does not prove a new requisition.
Wanted
Photoroom
Senior Backend Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris (hybrid). Application: If you are based in or around (Ile-de-France) Paris, are you willing to work from the Paris office at least a few days per week?
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay noteCompensation: 95k to 105k + Stock-Options/BSPCE. The shorthand does not explicitly establish currency, annual basis or gross cash-base classification; these are not inferred from the Paris address. ATS compensation tiers are empty.[46][5]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralfail
Payunknown
Product researchunknown
Gate notes
The role owns pricing architecture, subscription management and backend endpoint performance. Working at an AI imaging company does not make those responsibilities AI-central.[46]
Paris hybrid is stated and the application explicitly asks for Paris-office attendance at least a few days per week; geography passes on the combined role and application evidence.[46][47]
95k to 105k plus stock options/BSPCE is published, but currency, annual basis and gross cash-base classification are unconfirmed. A superficially attractive numerical range is not enough for a pay pass.[46][5]
The listing requests 5+ years in Django/Django REST Framework, relational databases at scale and SaaS subscription systems. Its application form renders.[46][47]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: own the pricing system across iOS, Android, web and API products; build backend features, optimize endpoints, monitor releases with Datadog and work with subscription providers such as Stripe/RevenueCat. A model, agent or evaluation project is not assigned to this role. Separate product diligence was not advanced.[46]
Source evidence notes (5)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: requisition c95a2585-fd0b-4d52-8b5e-b08acf53ce02; published September 28, 2026; Paris; no structured compensation tiers.
Wanted
Escape
Senior Research Engineer | Cybersecurity AI AppSec
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Location: Paris (75002) | Hybrid (2 days remote/week)
Salary evidence
No published figures
Equity
Yes
Pay noteSignificant equity (Stock Options/BSPCE) is offered, but the current role page has no numerical annual cash-base range and current ATS compensation tiers are empty. An older selection is not evidence of today's pay floor.[48][8]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Current annual gross cash-base floor is unknown: the opened listing has no numerical salary range and the current official ATS record has empty compensation tiers. Explicit equity does not replace a cash floor.[48][8]
This exact requisition was previously selected on October 3 and October 5 according to supplied selection history. It is a recheck, not a newly discovered recommendation; older qualification is not reused.
The current listing documents senior hands-on agentic scanning, LLM-driven security testing and production detection logic, with Paris hybrid attendance. Its application form is reachable.[48][49]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: agentic application scanning, vulnerability detection, LLM/structured-reasoning workflows and distributed orchestration of concurrent scans. Python, Go, Kubernetes/EKS, Playwright, Kafka and observability tools are named; 4+ years in backend engineering, security research or applied R&D are requested. Named internal project unknown. Separate product diligence was not advanced.[48]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteCurrent metadata summary: requisition 2384716d-587f-4ff2-a7e9-dd8557e49920; published September 17, 2026; Paris Office, Hybrid; compensationTierSummary null and compensationTiers empty. No current numerical salary floor is established.
Pay noteCompetitive compensation + equity — everyone shares in our success. No numerical annual cash-base range is published in the listing or employer ATS compensation fields.[52][19]
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
No numerical annual gross cash-base floor; equity alone cannot satisfy the compensation gate. This is missing compensation evidence, not a below-minimum published salary.[52][19]
France remote location, senior engineering level and AI-tool/agent evaluation and workflow construction are documented. Seven-plus years of software engineering and prior team-level AI-assisted development practice are requested.[52]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: XTM Foundation supports OpenCTI, OpenAEV and OpenGRC engineering. Duties include measured AI coding-tool/agent experiments, code-quality guardrails, CI agent workflows for dependency updates, flaky-test triage and review assistance, shared MCP servers and prompt libraries, and team enablement. This is an internal engineering-platform remit, not a guaranteed assignment to a customer-facing AI model. Separate product diligence was not advanced.[52]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteRun structured experiments to evaluate new AI coding tools and agents (Cursor, GitHub Copilot, Cline, MCP servers, code generation agents…), measure their real impact on velocity and quality, and translate findings into actionable recommendations.
Evidence noteMetadata summary: requisition 078b3281-6e64-4c52-82c9-c20a73dcb2fa, published October 9, 2026; France and four additional European locations; Remote; no compensation tiers.
Evidence noteWhat are your salary expectations for this role ?
Wanted
AGON AI
Research Scientist — Visuotactile Robot Learning
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Hybrid · Paris, France (Hybrid)
Salary evidence
Known annual EUR base
Equity
Yes
Pay note€85,000 – €115,000 / year + BSPCE / Equity. The careers page describes its ranges as base salary, but gross compensation is not explicitly stated. No unconditional pay pass is recorded.[37]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Paris Hybrid has no explicit office-attendance policy; France-eligible fully remote work is not offered for this particular role.[37]
The role summary and application modal do not establish senior/staff entry requirements. A Research Scientist title alone is insufficient.[37]
Annual base range and equity are documented, but gross classification remains unconfirmed. Other hard gates already prevent qualification.[37]
The role-specific application modal opened through its Apply / Details button and displayed five fields. No distinct permanent requisition URL or ATS identifier was exposed.[37]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer role-summary: foundation policies, action-chunking diffusion architectures identified as ACT, FARM and RDP, and visuotactile representations for contact-rich physical tasks. The application repeats this short summary rather than giving detailed seniority requirements or a narrower assigned project. Separate product diligence was not advanced.[37]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteDevelop novel foundation policies, action-chunking diffusion architectures (ACT, FARM, RDP), and visuotactile representations that generalize across contact-rich physical tasks.
Evidence noteEvery position — from hourly on-site teleoperators to principal hardware engineers and research scientists — states the exact base salary or hourly range in Euros upfront before you apply.
Evidence note€85,000 – €115,000 / year + BSPCE / Equity · Hybrid · Paris, France (Hybrid)
Wanted
Nango
Forward Deployed Engineer (Founding)
Rejected
Report date
Evidence checked
Posted
Work location
Remote — This is a remote role open to candidates based in North America & Europe
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay noteSenior: $120K – $200K • 0.05% – 0.15%. ATS identifies annual Salary; gross cash-base classification is not explicit. Original USD retained; no EUR conversion or eligibility claim made.[54][7]
Recorded gates (as checked on the report date)
Listing openpass
Locationunknown
AI-centralfail
Payunknown
Product researchunknown
Gate notes
AI-primary work is not established: responsibilities are customer integrations, implementation/platform debugging, pairing and general full-stack product improvements. Serving AI applications does not itself establish a primary AI-system engineering assignment.[54]
Remote North America/Europe does not explicitly establish France applicant/payroll eligibility.[54][55]
No positive pay decision: gross annual base and a dated EUR comparison were not established after the scope/location rejection.[54][7]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: API access and product integrations for agents and applications. This hire helps customers from technical evaluation to production, builds integrations, resolves implementation/platform issues and ships full-stack improvements. AI mechanism or named model-building project for this hire is not documented. Separate product diligence was not advanced.[54]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Unknown — No current official role-location evidence: listing and application display Job not found.
Salary evidence
No published figures
Equity
Unknown
Pay noteNo current readable official compensation evidence. Old aggregator pay ranges were not substituted for employer verification.
Recorded gates (as checked on the report date)
Listing openfail
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Official listing and application both render Job not found; no application form is available. HTTP 200 alone did not establish an open role.[44][51]
The requisition is absent from the retrieved current Docker board. Discovery title retained solely to identify the excluded lead; old salary, duties and location were not verified as current.[5]
Evidence noteBoard-inspection summary: requisition ebc9eb47-4d48-4e6c-82db-b38c7c31e826 absent from the retrieved board. The board does not establish this missing role's current salary or location.
Wanted
Mendo
Senior AI Engineer (unavailable)
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — Archived description only: Paris, Châtelet – Hybrid (2 days on-site / 3 days remote). Current employer page says the offer is unavailable.[56][57]
Salary evidence
No published figures
Equity
Yes
Pay noteConflicting archived wording: €75,000–€90,000 gross annual base + BSPCE, versus €80,000–€95,000 base + BSPCE elsewhere in the same description. No authoritative current lower bound selected; this is not a current open offer.[56]
Recorded gates (as checked on the report date)
Listing openfail
Locationunknown
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
Employer Teamtailor page returns HTTP 410 and says the offer is no longer available. The application endpoint returns 404 without a role application form.[57][58]
Official WTTJ API marks reference MENDO_7oeRWVL / b5f9d677-2174-490c-abcf-8fe655de3427 archived on September 12, 2026 and links employer requisition 6979783. Search visibility is not proof of an active opening.[56]
Archived salary floors conflict internally; choosing the higher range would improperly manufacture pay eligibility. Archived attendance/duties do not establish today's live conditions.[56]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteArchived employer-description summary only: on-device small language models, fine-tuning/RAG, evaluation harnesses and AI adoption/analytics features. The current employer page does not expose an active assignment. No current product/project diligence was advanced.[56][57]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence note€75,000–€90,000 gross annual base + BSPCE (equity), depending on level
Wanted
Joko
AI Software Engineer (senior/staff level)
Rejected
Report date
Evidence checked
Posted
Work location
Remote — This position is remote-friendly (France and Spain). Flexible remote : If you live in Paris, you can choose to work from our office or from home with no constraints.
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay note€76.3K – €174.6K • Offers Equity. Employer ATS: Salary, 1 YEAR. Fixed gross cash-base label is not explicit. The employer says final compensation may sit at, above, or below the published range.
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Published salary floor €76,300 is below €80,000; the €174,600 ceiling cannot satisfy the lower-bound rule.[55][37]
Final compensation is not a guaranteed published floor: the employer says it may fall below the range.[55]
France remote eligibility, explicit senior/staff entry and agent engineering are documented; the compensation gate remains decisive.[55]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: Joko AI is a shopping agent. Duties cover agent architecture, evaluations and regression monitoring, web/vector search tools, catalog ingestion and enrichment, personalization, model fine-tuning and production latency/cost/reliability. Five-plus years of software engineering and senior/staff ownership are requested. Separate product evidence was not advanced after the pay rejection.[55]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: Salary €76,300–€174,600; interval 1 YEAR; equity separately offered; published March 25, 2026. This is salary metadata, not an independently confirmed fixed gross base.
Evidence noteWhat do you think is the hardest part about building an AI shopping assistant that users can truly rely on?
Wanted
Inato
Senior AI Engineer
Rejected
Report date
Evidence checked
Posted
Work location
Remote — PARIS OFFICE or REMOTE FROM FRANCE; Remote-first philosophy & flexible hours
Salary evidence
Numeric range, annual basis unconfirmed
Equity
Yes
Pay note€70K – €95K • Offers Equity. Employer ATS: Salary, 1 YEAR, with equity separately offered. Fixed gross cash-base label is not explicit.
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payfail
Product researchunknown
Gate notes
Published annual salary floor €70,000 is below the €80,000-plus-equity gate; the upper bound is irrelevant to minimum eligibility.[57][36]
France remote location and senior AI engineering are documented. The application additionally requires France residence/tax conditions and full work authorization, plus strong English and French assessed during recruitment.[57][58]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: clinical-trial technology. Responsibilities include multimodal model orchestration, RAG over medical documents and patient histories, recruitment/site-selection agents and production pipelines. Example projects are a reasoning-model/visual-RAG prescreening tool and a site-selection agent coupled with mathematical optimization; neither is a guaranteed assignment. Separate product diligence was not advanced after the pay rejection.[57]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteMetadata summary: Salary €70,000–€95,000; interval 1 YEAR; equity separately offered; published July 30, 2026. Location: PARIS OFFICE or REMOTE FROM FRANCE.
Evidence noteThis role requires strong proficiency in both English and French, and both languages will be assessed during the recruitment process. Are you comfortable working fluently in both?
Wanted
OpenAI
Applied AI Engineer, Startups (Codex)
Near-miss
Report date
Evidence checked
Posted
Work location
Hybrid — Paris, France; Hybrid. Application: Are you able to work from a local office three days per week?
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical salary range or explicit equity offer appears in this Paris listing or its employer ATS compensation fields. Unknown pay is not a below-threshold salary.
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralunknown
Payunknown
Product researchunknown
Gate notes
No published numerical France annual cash-base lower bound; compensation cannot be verified against the gate.[59]
Paris attendance is explicit in the application: three office days per week.[60]
Senior/staff entry is not explicit. Presentations and community engagement are central; AI-system engineering as the primary assignment remains unverified, despite documented coding, workflow integrations and reliability evaluation.[59]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: the Go To Market team supports startup adoption of Codex through technical talks, demos, workshops and hackathons, plus selected customer workflows, integrations and automations. Reusable code and feedback to Product/Research/Engineering are part of the remit; this is not evidence of a dedicated model-building assignment. Separate product diligence was not advanced.[59]
Source evidence notes (3)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteAre you able to work from a local office three days per week?
Wanted
Owkin
Scientist II Machine Learning Engineer (LLM & GenAI Specialist)
Near-miss
Report date
Evidence checked
Posted
Work location
Remote — Position is based in our Paris offices or remotely in France, UK, or Germany.
Salary evidence
No published figures
Equity
Unknown
Pay noteNo numerical salary range or explicit equity offer appears in this listing or its employer ATS compensation fields. Unknown compensation is not insufficient published compensation.
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
No published numerical annual cash-base floor.[61]
Remote work in France is explicit in the description, despite Hybrid metadata.[61]
Five-plus years of Python software development, production deep-learning deployment and mentoring are requested. Scientist II is not explicitly mapped to senior/staff entry; that level remains unverified.[61]
Research lead — opening a listing is not a recommendation.
Company & role evidence notes
Role/product noteEmployer-listing summary: biomedical LLM/GenAI engineering bridging research and production. Duties include ML libraries and research workflows, experiment reproducibility, lifecycle management, research infrastructure and model prototyping. Requirements include PyTorch/TensorFlow, Transformers, Hugging Face, agents/RAG, cloud/SageMaker and distributed training. The company describes K Pro, but assignment specifically to K Pro is not established. Separate product diligence was not advanced.[61]
Source evidence notes (4)
Evidence notes are recorded research notes and may paraphrase the source; they are not guaranteed to be verbatim quotations.
Evidence noteExperience: 5+ years of professional experience in software development with Python and a proven track record of deploying deep learning models into production.
Evidence noteMetadata summary: requisition df08e875-432e-420f-ac18-543c07e0ee1b published July 15, 2026; Hybrid metadata; no compensation tiers provided. The description separately permits remote work in France.
Evidence noteFrance, Germany, Spain, or Italy — remote within these hiring locations; No numerical annual cash-base range or explicit equity offer published.
Wanted
Spiko
AI & data lead
Rejected
Report date
Evidence checked
Posted
Work location
Hybrid — up to 2 days per week and one full remote week per month
Salary evidence
No published figures
Equity
Yes
Pay noteNot published
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Competitive package depending on experience + Stock Options; no numerical annual cash-base floor.
Evidence noteup to 2 days per week and one full remote week per month; Competitive package depending on experience + Stock Options; no numerical annual cash-base floor.
Wanted
Runware
Staff Software Engineer, Serverless
Rejected
Report date
Evidence checked
Posted
Work location
Remote — Remote; France explicitly listed
Salary evidence
No published figures
Equity
Yes
Pay noteNot published
Recorded gates (as checked on the report date)
Listing openpass
Locationpass
AI-centralpass
Payunknown
Product researchunknown
Gate notes
Stock options are explicit, but no numerical annual cash-base floor is published.
Posting-date sources differ: main employer publication 29 April 2026; France-specific Workable version 26 August 2026. Neither is an October posting.
Evidence noteThis role can be based remotely or from one of our office locations listed on the job description; Published compensation covers Canada and the USA; no France-applicable annual cash-base floor.