Portrait
drop in · 3:4
About

Yang Deng

Born in China, trained in optical engineering at Rochester, did a Ph.D. at Duke on ML for metamaterials, now building simulation infrastructure in Boston.

My path into machine learning went through physics. I started in optical engineering — lenses, light, the mathematics of how electromagnetic fields behave in structured media. Ph.D. at Duke under Willie J. Padilla was where it all collapsed into one question: when the design space is enormous and the simulations are expensive, how do you find good designs fast?

That question turned out to be bigger than metamaterials. Surrogates, inverse design, Bayesian optimization, active learning — each a different way of stretching a simulation budget further. My thesis pushed a five-orders-of-magnitude speedup on one benchmark. My best papers are the ones that say clearly what would generalize and what wouldn't.

At KronosAI I get to build the infrastructure for that at production scale — connecting physics simulations to AI interfaces, writing the benchmarks that keep the agents honest. I joined as the 2nd engineer because the chance to shape something from near-zero matters more to me than most things.

Where I've been

2025—now

Member of Technical Staff

KronosAI

2nd engineer. End-to-end product infrastructure; GPU-accelerated simulation kernels (RCWA/FDTD/FDFD); evaluation benchmarks for physics-oriented agent workflows.

Boston
2024—2025

Software Engineer · CEM & Photonics

Metalenz

Fabrication-tolerant design pipelines; Slurm + AWS; reduced runtime 20% via multi-proc.

Boston
2019—2024

Ph.D. · Electrical & Computer Engineering

Duke University

Thesis: Machine Learning for Next Generation Metamaterials. Advisor: Dr. Willie J. Padilla.

Durham, NC
2014—2018

B.S. · Optical Engineering

University of Rochester
Rochester, NY
the full version, as a PDFDownload CV

Questions I get asked

I work on the layer that connects high-fidelity physics simulations to LLM-based agents — benchmark design, verification harnesses, GPU kernel integration. I joined as the 2nd engineer and helped bring on the 3rd.

A few things, not in any order

Something from Duke
lab · 2023
Somewhere in Boston
move · 2024
A painting I like
reference
Whatever matters
drop in