- 5+ years software/ML engineering with production systems; strong Python
- 2+ years building LLM applications in production not just prototypes
- Robust hands-on experience with LangGraph or similar agent frameworks (CrewAI, AutoGen,
OpenAI Agents SDK, Claude Agent SDK) — has built and operated production agentic systems, not just tutorials
- Deep grasp of agentic patterns: tool calling, state management, memory architectures, multi-agent
orchestration, human-in-the-loop design
- RAG experience: embeddings, vector DBs, retrieval evaluation
- Experience evaluating LLM systems — building eval datasets, automated scoring, and regression
testing for prompt and model changes
- Fluent with AI-assisted development (Claude Code, Codex, Cursor, or similar) — uses these tools
agentically to multiply output, not just for autocomplete