Ji Won (pronounced G1) is a Machine Learning Scientist in the Prescient Design team at Genentech. She would describe herself as a pragmatic Bayesian — her current research probes hierarchical, sparsity-inducing structures in the data that can inform sampling, optimization, inference, and adaptive decision-making. She focuses on developing algorithms in MCMC sampling, posterior inference, and Bayesian optimization inspired by challenges in molecular design.
In her past life as an astrophysicist, she studied gravitational lensing using hierarchical Bayesian models to understand the origin and evolution of the Universe. She interned at NASA Ames and the Center for Computational Astrophysics at the Flatiron Institute while pursuing her Ph.D. in Physics at Stanford University, which she completed in 2022 under the supervision of Phil Marshall and Aaron Roodman. She holds B.S. degrees in Mathematics and Physics from Duke University (2017).