AI/ML Engineer at Databricks

Developed Python-based multi-agent AI workflows using RAG, LangGraph, vector search, and MLflow, improving response accuracy by 32% and supporting enterprise knowledge retrieval for 10,000+ active users. Optimized Python retrieval pipelines using semantic search, embedding optimization, and reranking strategies, reducing response latency by 38% and lowering inference costs by 28%. Built automated Python LLM evaluation and synthetic data generation pipelines with MLflow, reducing hallucinations by 27% and accelerating validation cycles by 40%. Engineered enterprise-grade agent applications using Python, LangGraph, MLflow, DataBricks Vector Search, Delta Lake, Unity Catalog, and foundation models for governed retrieval, orchestration, evaluation, and monitoring. Designed advanced RAG architectures integrating hybrid search, tool calling, and foundation models to enable contextual reasoning and enterprise knowledge discovery. Implemented observability frameworks with MLflow tracking, eval