

AI engineer and data scientist specializing in Python with expertise in LLM applications, agent systems, and applied machine learning. Proven track record in developing MCP tools, evaluation pipelines, and AI gym environments for testing reasoning and planning capabilities. Proficient in backend development, data workflows, and model-driven applications.
AI workflows
Model evaluation
Data warehousing
Languages: Python, SQL
Infra / Dev Tools: Docker, Git, Azure, MQTT, MongoDB
Visualization / Apps: Streamlit, Power BI
LightRAG Retrieval System
Built a retrieval augmented generation system using LightRAG for document search and grounded question answering over custom data sources. Developed ingestion, chunking, indexing, and retrieval workflows to improve context relevance and answer quality. Combined retrieval with LLM generation to support domain specific information access.
LangChain Chess Analyzer
Built an interactive chess analysis application using Streamlit, Stockfish, and LangChain. Identified blunders, missed tactics, and evaluation swings using centipawn based engine analysis. Generated phase wise game summaries from PGN data using LLM based insights for opening, middlegame, and endgame review.