BSES Rajdhani Power Limited
Built an on-premises RAG system that checks transformer datasheets against IS standards and past approvals, returning pass/fail verdicts with exact clause and page citations. Cut design review time by over 50% per vendor.

I learn by
breaking & building.
But mostly by breaking.
I'm an Electronics and Communication Engineering student at Manipal Institute of Technology, graduating in 2027. I spend most of my time building AI systems that don't phone home - retrieval pipelines, on-device inference, and the unglamorous plumbing that makes them auditable.
My work has taken me through power infrastructure at BSES Rajdhani, internal automations at the National Stock Exchange, and a cross-platform knowledge app I'm building on Rust and llama.cpp. The through-line is the same: get the model close to the data, keep the citations honest, and make it fast enough that people actually use it.
Education
Manipal Institute of Technology
B.Tech - Electronics and Communication Engineering
Graduating - Jun 2027
Based in
New Delhi / Manipal, India
Currently
Building Tesseract - a local-first knowledge system with a fully on-device AI pipeline.
Built an on-premises RAG system that checks transformer datasheets against IS standards and past approvals, returning pass/fail verdicts with exact clause and page citations. Cut design review time by over 50% per vendor.
Built a repo analysis tool that visualises code flow and dependencies with locally hosted LLM summarisation, plus CI/CD automations that detect and resolve merge conflicts.
Launched and led a podcast segment that took monthly viewership from 12k to 50k in under two months. Produced the channel's top video at 41k+ views.
The ones I'd put in front of you first. Each of them keeps the work on the machine it runs on, and each had to earn that constraint rather than inherit it.
Talk for two minutes about someone you just met. The phone transcribes it, extracts a structured profile, and never sends a byte anywhere for inference.

A local-first knowledge system that runs its entire AI pipeline on-device. No cloud calls, ever.

Where I spend most of my time. On-device inference and retrieval that can cite its sources.
Python and TypeScript daily; Rust for anything that has to be fast and local.
Full-stack web, plus cross-platform desktop and mobile from one codebase.
Enough ops to ship and keep things running.
Exploring roles in AI engineering for 2027. Always up for a conversation about local-first systems.
© 2026 Nikhil Sah
Built with Next.js, GSAP & Three.js