AI engineer building LLM systems that hold up in production.

I design and build LLM systems end to end: agentic workflows, document intelligence, retrieval, and the evaluation that shows whether any of it works. I care about what decides success in production: cost, latency, observability and knowing when an answer is wrong.

I currently work at Stozia, and I'm most interested in data systems for agents: retrieval over documents that change, and evaluating agents at scale. On the side I write about what I learn and build tools like these

Find me on GitHub, LinkedIn and X, or write to me at bprathamesh2003@gmail.com.