Description:

Join Haven Safety AI to build the enterprise learning intelligence layer for safety in high-risk organizations.
Work on AI and LLM engineering to connect evidence base and knowledge graph to product experiences for investigators and safety leaders.

Requirements:

Experience in AI or ML engineering with production LLM systems.
Hands-on experience with multi-step, tool-calling workflows using frameworks like LangGraph or LangChain.
Strong evaluation discipline for LLMs, including datasets and human review loops.
Ability to assemble context for LLMs and make informed retrieval judgments.
Proven track record of owning production systems from deployment to monitoring.
Comfort with various model providers and understanding tradeoffs in quality and operational risk.
Familiarity with Neo4j and Cypher for graph reasoning is preferred.
Strong Python skills, including FastAPI and asynchronous services.

Engage in meaningful reasoning problems where correctness is crucial.
Work in an evaluation-first culture that emphasizes measurable quality.
Directly impact safety teams in various industries with visible customer feedback.
Collaborate closely with the CTO and make significant technical decisions in a small team environment.

Skills