Rakshan Shetty, full stack, backend and GenAI engineer
I build web applications end to end in React, Next.js, Node.js and Python, design the REST APIs and data models behind them, and add LLM features such as RAG pipelines and AI agents with LangChain, LangGraph and the Gemini and OpenAI APIs. Much of my work is about shipping software faster with AI coding agents: running several agents in parallel, coordinating them, handing plans between them and giving them memory they can trust.
Open source
Orchestration for AI coding agents. Claude Code plans and reviews while Cursor, Codex, Gemini and free models build in parallel: 10 to 15 agents on one codebase without conflicts, one shared knowledge base, team mode and one-click project setup. TypeScript, 641 tests, MIT. Case study. Source on GitHub.
Evidence-checked memory for AI coding agents. Every fact quotes the code that proves it and is served only while that code still exists; when nothing holds, recall answers "I don't know". CLI and MCP server for Claude Code, Cursor and Codex. MIT. Source on GitHub.
Skills
TypeScript, JavaScript, Python, React, Next.js, Node.js, Express, Django, Flask, REST APIs, PostgreSQL, MongoDB, MySQL, Redis, Docker, AWS, GitHub Actions, RAG, LangChain, LangGraph, LangSmith, Gemini API, OpenAI API, MCP, AI agents, multi-agent orchestration, TensorFlow, scikit-learn.
Contact
rakshanshetty2003@gmail.com, GitHub, LinkedIn, Resume (PDF).