I'm a Quantum Architect at PsiQuantum Corp. in Palo Alto since 2022, where I contribute across the entire stack: I build classical and quantum simulation tools, run photonic experiments, invent calibration algorithms, and contribute to data analysis automation pipelines. At PsiQ, I specialize in quantum photonic single-photon switch multiplexers.
Previously: I am a Stanford lifer, getting my BS in Physics (Honors, 2015) and masters in CS (AI, 2016) at Stanford University before my PhD in photonic computing (EE, 2022). From 2021 to 2022, I was an AI resident at Google X, Moonshot Factory running large-scale constrained optimizations for photonic inverse design. Prior to my PhD, I was a data scientist and engineer at Stella AI, a now-acquired AI startup in the recruiting space where I developed Bayesian job-candidate recommendation models.
I'm interested in photonics, computing hardware, physics, deep learning, generative AI, and image processing. My PhD was, in short, "photonics for AI and AI for photonics." I built a novel analog optical matrix multiplication chip and integrated it with my custom optical probing station with a movable stage to enable the first ever neural net backpropaagation and cryptocurrency demonstrations on photonic circuits. These are included in my defense slides and the highlighted papers below.
The first ever demonstration of optical backpropagation in photonic neural networks, paving the way for a new future for machine learning and ultra-fast photonic device error correction.
Exploring the matrix optimization of unitary networks for different sizes and architectures, relevant to both photonic circuits and machine learning applications.
A new mathematical framework for representing unitary and orthogonal matrices using binary tree cascades, useful for low-rank hardware implementations and representations.
Updated as of May 2024. Template credit: Jon Barron.
