Shirley Chan

To usher in a new class of machine learning for scientific data, building models that can leverage shared concepts across disciplines. We aim to develop, train, and release such foundation models for use by researchers worldwide. We build a specialized AI agent that elucidates organic molecular structure from raw experimental NMR data, framing the problem as a constrained search over the chemical space. An LLM agent to execute protein-design tasks with RosettaScripts. A physics foundation model finetuned on idealized simulations transfers zero-shot to laboratory turbulence, crossing a decades-old sim-experiment gap in the mixing growth rate.