About The Role:
As a Senior AI Engineer (GenAI), you will own production GenAI systems end-to-end, from the eval set, to the agent design, to the FastAPI handler running in production, to the dashboard you check the next morning.
You'll work directly on systems like our customer-support agent, which today drafts replies to real customers and CS reps in a regulated insurance setting ,no toy demos, no handoff to a separate engineering team.
This is a high-ownership role where your work directly influences our products, operations, and long-term AI strategy.
What You’ll Do:
- Design, build, and ship agentic workflows end-to-end, agent logic, RAG, FastAPI service, deployment, monitoring
- Build and own evaluation pipelines: golden datasets, LLM-as-judge calibration, regression tests, distribution-shift detection
- Productionize RAG pipelines grounded in insurance-specific data
- Integrate LLM agents with policy/claims systems and internal tools
- Own observability, guardrails, safety, and error handling for systems real users
Requirements
- 6+ years in AI/ML, Data Science, or applied engineering; 1-3 years shipping production GenAI applications (LLMs, agents, RAG) — not demos
- Master’s or PhD in CS, ML, Statistics, Engineering, or related , or equivalent practical experience
- Hands-on with multi-agent orchestration, tool use, and context-managed conversational systems (OpenAI Agents SDK, LangGraph, or similar)
- Deep expertise in RAG, vector databases, embedding selection, and retrieval optimization
- Strong Python + ML foundations (PyTorch, scikit-learn, Hugging Face)
Why Honeycomb
- Build foundational AI systems in a sector ripe for innovation and modernization
- See your work deployed quickly with immediate customer and business impact
- Collaborate directly with product, engineering, and executive teams—minimal bureaucracy
- High autonomy, fast development cycles, and meaningful equity ownership
- Be part of a mission-driven company redefining what insurance technology can be
