I've built production-ready Generative AI, RAG, and agentic AI applications at MetroLabs using Python, OpenAI GPT, Gemini, Claude, LangChain, LangGraph, and Vertex AI.
I develop end-to-end document and knowledge-discovery systems, from ingestion, parsing, chunking, and embeddings through vector retrieval and response generation. My work includes semantic and hybrid search with ChromaDB, Pinecone, FAISS, OpenAI Embeddings, and enterprise data from BigQuery, APIs, documents, and cloud repositories.
I've exposed AI capabilities through Python REST APIs and microservices, built multi-step agent workflows with tool/function calling and MCP-based tools, and implemented validation, human-in-the-loop controls, logging, governance, and audit controls.
Previously, I built AI data-processing and ETL/ELT platforms at Clarivate Analytics, improving information retrieval accuracy by approximately 45% and reducing employee search effort by more than 60%. My data engineering background at Capgemini and Unilever includes Python, SQL, PySpark, Spark SQL, Airflow, data quality, lineage, warehouses, and data lakes.
