I'm Michael — a chemistry & computer-science student at Harvey Mudd, working in the Zhuang group on the structural heterogeneity of supercooled water: large MD simulations, order-parameter embeddings, clustering into two liquid-like states.

Outside the lab I build tools at the seam of atoms and algorithms — agentic spectroscopy reasoning, sparse autoencoders on RL-trained models, GNN pipelines for MS/MS fragmentation. I care about systems you can hold in your head, and results that survive being looked at twice.

Most of what I make is here: code, preprints, and a few unfinished sentences I'm still rewriting.

I'm a first-year student at Harvey Mudd College studying chemistry and computer science. Currently, I spend my time:

• →
  Researching Atoms — running large-scale MD simulations under Prof. Bilin Zhuang, studying structural heterogeneity in supercooled water with UMAP, HDBSCAN, and GMM clustering.
• →
  Interpreting RL — researching mechanistic interpretability of reinforcement learning, training Sparse Autoencoders and tracking representational drift.
• →
  Exploring GNNs — actively learning graph neural networks, currently working on migrating MS/MS fragmentation prediction models (ICEBERG) from DGL to PyTorch Geometric.