Suryash Malviya

I build the Unity electromagnetism simulations Cornell uses to teach E&M — invisible fields turned into things students can grab, bend, and break.

I explore quantum error correction with modern ML tooling — GPU-first experiment pipelines, learning-based QEC ideas, and performance work that made it into Apple's MLX.

The work I've contributed back to the world: a merged optimization in Apple's MLX framework and peer-reviewed research on algorithm pedagogy.

I spent a research summer making our campus telescopes photograph the sky — CCD instrumentation, filters, and processing pipelines that made astrophotography something any student can do.

I run the hands-on side of my work at the Ithaca College Makerspace — CAD, 3D printing, and firmware, from World Science Festival exhibit parts to resurrecting a 2012 printer with modern firmware.

My CS research spans two threads: how to teach dynamic programming (published in JCSC with Prof. Ali Erkan) and how LLMs behave on GPUs (ongoing with Prof. Venkata Govindarajan — the work behind my MLX contribution).