At Tata Consultancy Services, I build production-oriented RAG systems that ground answers in internal documents using LangChain, ChromaDB, hybrid retrieval, and metadata filtering.
I've reduced knowledge search time by 30% and lowered average LLM latency from 2.1 seconds to 1.3 seconds through inference profiling and semantic query caching.
I also engineered LangGraph-based multi-agent workflows for planning, retrieval, and generation, cutting research turnaround time by 25%. My AI-Orchestrated Research System uses eight agents to turn 1–2 hours of research into citation-backed reports in under 30 minutes.
Earlier, I built React interfaces and REST APIs at Tata Consultancy Services, reducing UI latency by 20%. I bring Python, JavaScript, SQL, and full-stack delivery experience to AI-powered products, including ResumeOS, an AI resume optimization platform.
