About Me
I am a Master’s student in Data Science at ETH Zurich, where I focus on efficient machine learning for scientific computing. I previously earned a B.S. in Computational Physics with a minor in Mathematics from UC San Diego.
I work at the intersection of physics, mathematics, and machine learning. My goal is to build efficient machine learning models grounded in the physical and mathematical principles that govern both the world and the tasks they are designed to solve.
My research centers on model compression and efficient deep learning for scientific problems. I work on quantization-aware training, pruning, and knowledge distillation to make deep learning models faster and lighter without sacrificing performance.
Notable Papers
Neural Architecture Codesign for Fast Physics Applications
This paper presents a new hardware-aware neural architecture search pipeline for low latency deployment
wa-hls4ml: Benchmark and Surrogate Models for hls4ml Resource Estimation
Benchmark and dataset of over 680,000 synthesized neural networks to evaluate GNN- and transformer-based surrogate models predicting FPGA resource usage and latency