Meet Jadhav

I've built an LLM/RAG evaluation platform that measures answer quality, hallucination, faithfulness, retrieval precision and recall, citation correctness, latency, and cost. I implemented quality gates, metric-aware regression detection, async batch evaluation, LLM-as-a-judge scoring, and a dashboard for run history and baseline comparison.

I also designed a full-stack RAG knowledge assistant using FastAPI, PostgreSQL/pgvector, and React/TypeScript to provide document-grounded answers with source citations. My work includes authentication, semantic search, document processing, observability, rate limiting, migrations, Docker deployment, CI workflows, and automated testing.

Alongside AI engineering projects, I collaborate with production and quality teams to troubleshoot process issues, analyze data, maintain documentation, and support manufacturing targets.