We're hiring an Agent Engineer to design, ship, and harden production AI agents end-to-end.
What you'll do - Design and ship production AI agents — orchestration, composition, state management. - Build eval and reliability tooling for agents running thousands of times a day. - Contribute to institutional-knowledge capture and retrieval systems (tribal knowledge, RAG, meeting-notes agents). ## What we look for - Hands-on experience shipping agents in production — professional, OSS, or personal all count. - Clear opinions on orchestration frameworks (LangGraph, AutoGen, CrewAI, Temporal, home-grown DAG executors). - Comfort reasoning about multi-agent composition, state shape, and failure modes. - Experience building eval stacks for agents at scale (offline sets, canary, shadow deploys, LLM-as-judge). - Concrete thinking — named tools, real tradeoffs, specific regressions. ## Interview topics (async, ~20 min, AI-led) 1. Agent systems you've built — orchestration approach, build-vs-buy, framework tradeoffs. 2. Composition & state — how agents trigger each other, state shape per hop, failure handling. 3. Evals & reliability — offline + online + canary layers, concrete regression stories. 4. Institutional knowledge — capture systems you've built or admired, retrieval gaps, where knowledge decays. ## After you accept the offer I'd love to schedule a brief live follow-up to dig into your async answers. Grab a time on my calendar : [ signals move to a live conversation with the team.