At IBM's AI Center of Excellence, I design and deliver production GenAI and RAG features. My retrieval work improved answer relevance by 40% and reduced API latency from 2.8 seconds to 1.6 seconds.
I also fine-tuned and evaluated IBM Granite models, built Python and FastAPI services handling 10K+ daily requests, and deployed containerized applications with Docker and Kubernetes. For my personal project MediMind AI, I built a multi-agent clinical research platform using LangGraph and agentic RAG.
