Summary
Joy Zhang is a research scientist specializing in kernel development and optimization for custom AI inference and training, currently contributing to PyTorch MTIA work at Meta. With an 11-year track record bridging PhD-level research at Cornell and production-focused internships, she brings deep expertise in numerical methods for thin-structure simulation, cloth and hair modeling, and large-scale ML systems. Her doctoral thesis and papers show a rare combination of physics-based animation and high-performance systems thinking, applied to real-world engines and ML kernels. Joy has moved smoothly between academia and industry—shipping simulation research at Meta and Adobe, contributing to game-engine rendering at Ubisoft, and building ML prototypes during undergrad research. She is comfortable across the stack from mathematical modeling and C++/GPU kernels to PyTorch integration, and often focuses on multi-scale problems where algorithmic insight unlocks practical performance gains. Based in Ithaca, she pairs rigorous Cornell training with hands-on production experience to make sophisticated research directly usable in industry.
11 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Cornell University
Bachelor of Mathematics Computer Science (Fine Arts option) Statistics Computational Mathematics, Bachelor of Mathematics Computer Science (Fine Arts option) Statistics Computational Mathematics at University of Waterloo
English, Chinese