At TCS, I build enterprise multi-agent QA platforms using LangGraph, FastAPI, Python, Anthropic Claude, and OpenAI models. I designed Supervisor-Orchestrator architectures that route requests across requirement analysis, JIRA, test case generation, test script generation, and execution agents.
I build RAG capabilities for JIRA, Confluence, PDFs, and Office documents through semantic chunking, embeddings, metadata filtering, vector search, and citations. I’ve also generated Playwright scripts, Page Objects, and Gherkin feature files from natural-language requirements, while implementing DeepEval and LangSmith evaluation for hallucination, faithfulness, answer relevancy, and regression testing.
My background combines machine learning proof-of-concepts with more than 11 years in software quality assurance and automation. I’ve led testing teams and delivery activities across airline, banking, and telecommunications projects, including test strategy, customer demos, go-lives, automation frameworks, and project ownership.
