Agentic AI Engineer

Design and ship specialist AI agents using memory, tools, data and multi-step reasoning. Build multi-agent orchestration, confidence evaluation, human-in-the-loop escalation, and workflows for sport scientist expertise. Create evaluation, observability, and regression testing for agent performance. The role focuses on multiple specialist agents collaborating to answer complex performance questions with grounded, calibrated recommendations.

Seeking a senior AI engineer with production agentic AI experience: must have shipped a production system using memory, tool use, multi-step reasoning. Experience with multi-agent orchestration (LangGraph, AutoGen, CrewAI) and confidence calibration (Platt scaling, isotonic regression, ECE). Strong Python, software engineering fundamentals. 5+ years applied ML/AI/software engineering. Preferred: RAG, fine-tuning, LLM observability, human-in-the-loop, Go, AWS, GraphQL, PostgreSQL.