AI/product engineer building observable agent systems, retrieval pipelines, and developer tools.
India (IST) · Available immediately for remote B2B AI/Product Engineering contracts
Portfolio · Resume · LinkedIn · Email
MCPHub — developer tooling for MCP
- Test MCP servers through schema-generated forms and inspect raw JSON-RPC over SSE, Streamable HTTP, and stdio.
- Track P50/P95/P99 latency, emit JSON/JUnit results for CI, and detect OAuth issues, tool poisoning, and schema drift.
Next.js TypeScript PostgreSQL Drizzle MCP SDK
AgentMesh — visual multi-agent orchestration
- Build sequential, conditional, and parallel agent workflows on a visual DAG canvas.
- Inspect typed WebSocket events, tool calls, state transitions, and human-in-the-loop approvals in real time.
Python FastAPI Next.js WebSockets PostgreSQL MCP
- AI Engineer Intern · Infradock.ai (Jan 2026–Present) — improved retrieval precision from 55% to 95%+ using clause-aware chunking and cross-encoder reranking; stress-tested across 500+ queries.
- Software Engineer Intern · Oldowan Innovations (Oct 2025–Jan 2026) — built Django matching APIs and Celery/Redis notification infrastructure for Zingles, a live consumer application.
- Software Engineer Intern · Yantram Medtech (Jul–Oct 2025) — built Node.js/Express clinical APIs with JWT/RBAC and reduced PostgreSQL and MongoDB query latency by 25% through indexing and query restructuring.
Python, TypeScript · FastAPI, Django, Node.js · LangGraph, LangChain, RAG, MCP · Next.js, React · PostgreSQL, Qdrant, pgvector, Redis · Docker, AWS
- NASA Space Apps Challenge 2025 — Regional Winner and Global Nominee
- Smart India Hackathon 2024 — Grand Finalist
B.Tech CSE (AI/ML), OIST Bhopal — 2026