Location: Guwahati, India (IST, UTC+5:30)

Remote: Yes - remote only. Full overlap with European hours and US-East mornings.

Willing to relocate: No

Technologies: TypeScript, Node.js, React, Next.js, PostgreSQL, Redis, MCP, OAuth/OIDC, JWT, AWS, Docker, CI/CD, XState, Yjs/Liveblocks, Vitest

Résumé/CV: https://drive.google.com/file/d/1ZM7rudeKxKgCMBckatiI49e321Q...

Portfolio: https://tusharnath.online

GitHub: https://github.com/tushar-nath

LinkedIn: https://linkedin.com/in/tusharnath10

Email: [email protected]

I build the code around agents: the tools they call, the auth around those tools, the async work behind them, and the interface that shows people what a run is doing.

Previously at Chronicle, I built its production MCP server end to end: 8 tools used by Claude and ChatGPT, WorkOS OAuth with JWT, async polling, and deployment on AWS ECS.

I also built Chronicle’s first public REST API, developer docs, a workspace-scoped API key system, and an npm integration that passed n8n’s strict-mode security review. On the product side, I worked on an AI-generation interface with streaming tool calls, cancel/abort mid-run, and live document editing over CRDTs. Public write-up: https://chroniclehq.com/blog/introducing-chronicle-mcp-api

I independently built Recall in about 2 weeks: a paid web product that turns notes into flashcards through an 8-tool MCP server, then schedules reviews using FSRS. I owned the product, backend, auth, billing, database, deployment, and launch. Product: https://recallapp.online Demo: https://www.youtube.com/watch?v=zPDFX9490Jo

I have around 4 years of experience and work best when I can take a rough problem through design, implementation, deployment, and the inevitable debugging after launch.

I’m looking for a remote backend, full-stack, product engineering, or agent-infrastructure role. Full-time and contract both work. Available immediately.