At Accenture India, I built and deployed agentic, RAG-based LLM microservices for a global enterprise client, using Python, FastAPI, and LangChain/LangGraph. These workflows combined retrieval, vector search, and tool-calling, and reduced decision-automation latency by 30%.
I also developed production ETL pipelines and AWS/GCP infrastructure, and created a shared GitLab CI/CD pipeline that cut deployment time by 80%. In a personal project, I built MINISEARCH, a distributed hybrid search engine combining BM25 ranking with vector similarity search.
