About
Hi, I'm Nanometers. I build things at the boundary of machine learning and systems. My work spans efficient inference, differentiable physics, and distributed computing.
I'm fascinated by the engineering challenges that emerge when you push models to their limits—how memory hierarchies, numerical precision, and communication patterns shape what is computationally feasible. Most of my recent work centers on making large-scale inference cheaper and faster without sacrificing quality.
Outside of research, I write about the mathematics and systems intuitions that underpin modern ML. I believe clear exposition is one of the most under-valued contributions a researcher can make.
When I'm not staring at GPU profiles, you'll find me reading about ergodic theory, tinkering with compilers, or brewing pour-over coffee.
Timeline
B.S. Computer Science
Graduated with honors, focusing on algorithms and numerical computing.
Research Intern
Worked on distributed systems and GPU kernel optimization at a leading research lab.
M.S. Machine Learning
Thesis on efficient inference methods for large language models.
Published at NeurIPS, OSDI
First-author papers on mixture-of-experts architectures and distributed KV-cache systems.
Current: PhD Researcher
Investigating the intersection of systems design and large-scale machine learning.