LLM Project / Code Review

Bring your LLM project — we debug and improve it live. I build production conversational AI (RAG on pgvector, agentic API execution, tiered model routing on AWS Bedrock) and apply the same review lens to your code.

What we can cover in 60 minutes:

• RAG quality: chunking, retrieval scores, hallucination guards

• Agent/workflow architecture and tool design (incl. MCP)

• Prompt structure and model-tier cost optimization

• FastAPI/Python backend patterns for LLM apps

Come with your repo or architecture diagram. Leave with concrete fixes, not generic advice.