Purshottam Kumar

I’m a Software Engineer focused on building scalable backend systems and AI-driven applications, specializing in Generative AI and Retrieval-Augmented Generation (RAG). I bring a strong foundation in FastAPI, Node.js, React, and AWS, with a focus on clean architecture, performance, and reliability.

In my current role, I built a distributed backend platform using FastAPI, MongoDB, Redis, and Elasticsearch, improving workflow efficiency by 60%. I also implemented time-series forecasting with real-time, low-latency inference pipelines to support market-trend prediction and data-driven decisions, and optimized APIs and database queries to reduce end-to-end latency by 40% at scale.

I’ve delivered end-to-end AI assistant capabilities with a LangChain-powered RAG workflow using Pinecone, improving response accuracy by 30% and reducing query resolution time by 40%. I’ve also worked on authentication and scalable performance improvements as a Full Stack Developer Intern, plus analytics dashboards in Power BI.