Arjun Sahlot
I love computers. I’m studying them at UW–Madison. Last summer I worked on Locus and Summary Bot, along with a bunch of fun AI experiments in the lab, while holding an ML engineering position at Magic Hour. The previous summer, a software engineering position at MathWorks.
Most of what’s here started as a question I wanted a real answer to. Some of it became finished work. The rest is still being figured out.
Selected work
Diffusion Maze Solver
Pathfinding by denoising instead of search. A U-Net trained from scratch takes a 23×23 maze and a field of noise, then walks a thousand reverse diffusion steps until the noise resolves into a path from start to goal. It runs live in the browser.
Locus
A mobile workspace for working with AI. Each space pairs a conversation with a document the agent can edit directly, so the result of the work outlives the chat that produced it.
ChessBench
A standardized harness that asks language models to write chess engines, then plays the engines against each other. Roughly three thousand games of round-robin turn a subjective question into an Elo number.
A few experiments going, none of them ready to be looked at yet.
- GPT-2, but better·
- FrameFT experiments·
- notes/tools collection
Experience
- Magic Hour (Remote, from Oakland, Jun – Aug 2026)ML engineeringYC W242026
- MathWorks (Boston, Massachusetts, May – Aug 2025)Software engineering2025
- UW–Madison (Madison, Wisconsin, Graduating May 2027)B.S. Computer Science2024–present
Writing
Nothing published yet — coming soon.
Elsewhere
Email is the best way to reach me — iarjun.sahlot@gmail.com. I’m also on GitHub, X, and LinkedIn.