Enhancing Diffusion-Based Sampling with Molecular Collective Variables
Flow Matching for Accelerated Simulation of Atomic Transport in Crystalline Materials
Interpolation and Differentiation of Alchemical Degrees of Freedom in Machine Learning Interatomic Potentials
Transferable Learning of Reaction Pathways from Geometric Priors
Think while You Generate: Discrete Diffusion with Planned Denoising
Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolation
Research Experience
PhD Candidate at MIT DMSE & CCSE; Part-Time Student Researcher on the FAIR Chemistry team at Meta.
Education
PhD Candidate at MIT DMSE & CCSE, advised by Prof. Rafael Gómez-Bombarelli, currently in the third year.
Background
Studies how atoms move and rest, striving to turn their patterns into things built to last. Technically, works on ML interatomic potentials, free energy methods, and generative modeling to understand and design molecules and materials.