Recourse Health
AI Engineer, Decision Intelligence
San Francisco, United States · On-site · Full-time · Posted Jul 29, 2026
- Experience
- 5+ years engineering; meaningful recent production LLM or ML product experience required.
About the job
Own the AI decision-intelligence layer for healthcare arbitration disputes, including claim selection, offer amounts, argument generation, evaluation, and learning from outcomes.
What you'll do
- Build the engine for deciding which claims to contest and what amounts to offer, and generate dispute arguments.
- Create evaluation infrastructure and feedback loops that measure outcomes and improve future decisions.
- Design and explain the division between deterministic rules and model reasoning; build recommendations customers can understand.
What they ask for
- 5+ years of engineering experience, including meaningful recent work building production LLM-powered or ML-driven products.
- Experience building evaluations, such as golden datasets and offline tests, and measuring whether changes help.
- Ability to work across the stack and ship products end to end.
- Comfort with TypeScript or Python; strong judgment about where probabilistic reasoning is appropriate.
Nice to have
- Experience in regulated or security-sensitive environments, such as HIPAA, SOC 2, PCI, or financial controls.
- Healthcare data, claims, or an adversarial decision domain such as fraud, risk, pricing, or trading.
Tools
TypeScript, Next.js, Prisma, PostgreSQL, Google Cloud, Claude Code, Codex
AI focus
Build production LLM/ML decision systems for arbitration, including claim and offer recommendations, argument generation, evaluation, explainability, and learning from outcomes.
Model-written summary · the original posting is authoritative