Summary
Tim Hsu is a data scientist with nine years of applied research experience in geometric and graph-based machine learning for atomic and molecular structures, currently developing generative ML methods at Lawrence Livermore National Laboratory. His work bridges scientific AI and high-performance computing—spanning graph neural networks, generative models for atomic configurations, and distributed training across dozens of GPUs. Tim’s background includes a top-tier PhD in Materials Science from Carnegie Mellon, where he combined 3D microscopy, tomography, and C++ physics simulations on hundreds of CPU cores, reflecting strong cross-disciplinary fluency. He is adept at turning unstructured scientific data into predictive and generative models that accelerate materials discovery, and brings a practical track record of scaling research prototypes toward production-grade HPC workflows.
9 years of coding experience
4 years of employment as a software developer
University of Illinois Urbana-Champaign
PhD, Materials Science and Engineering, 3.96/4.00, PhD, Materials Science and Engineering, 3.96/4.00 at Carnegie Mellon University
Chinese, English, japanese (jlpt n2)