Mistral AI is hiring a Staff Staff LLM Engineer — Agents & tool use on the Applied research → prod team (Hybrid · Berlin, Germany). This is a hands-on role in LLM & generative AI—not a generic “AI enthusiast” post.
You will own LLM features, retrieval, agents, and evaluation end to end: problem framing, offline evaluation, safe rollout, and iteration from production signals. Success is measured with clear metrics (quality, latency, cost, or business KPIs)—not slide decks alone.
Our applied team ships agentic assistants for operations teams across the DACH region. We need a Staff engineer who can own architecture for multi-step workflows, safety layers, and multilingual rollouts.
Berlin hybrid; you will collaborate with research, infra, and compliance from day one.
What you will do
- Design planning, retrieval, and execution graphs with explicit human-approval gates.
- Lead postmortems for tool failures, prompt regressions, or data leakage near-misses.
- Establish engineering standards: tracing, structured logging, redaction, and secret hygiene.
- Coach senior ICs; review designs for cost, latency, and maintainability.
- Represent engineering in customer security reviews for agent features.
What we look for
- 8+ years production software with 3+ on LLM/agent systems at scale.
- Expert Python; deep knowledge of vector DBs, OpenAI-compatible APIs, orchestration frameworks.
- Written communication that scales across time zones (EN + DE professional fluency ideal).
- Evidence of cross-functional leadership without people-management requirement.
Nice to have
- Experience with regulated workloads (finance, healthcare adjacent).
- Contributions to OSS agent frameworks or retrieval papers.