Production RAG that holds up
Hybrid dense + sparse retrieval, reranking, streaming citations and a confidence gate that refuses to answer on weak evidence — shipped, not slide-ware.
Hire me
For companies and recruiters. I build RAG systems, fine-tune open models and ship multi-modal AI products to production, and I can own the full stack around them. Book a call below — the request reaches me on WhatsApp.
What I bring
Hybrid dense + sparse retrieval, reranking, streaming citations and a confidence gate that refuses to answer on weak evidence — shipped, not slide-ware.
QLoRA through Transformers, PEFT and TRL, live training telemetry, and side-by-side base-versus-tuned comparison before anything is called better.
FastAPI and Next.js, PostgreSQL and vector stores, Docker and CI/CD with logging and metrics from day one. Five services ran this way in production.
Proof of work