I graduated from Dickinson College in May 2026 with a B.S. in Computer Science and Mathematics and a minor in Economics. Over four years, my work spanned a wider-than-usual range: deep learning models that outperform physics-based methods at predicting energy localization in nonlinear lattices; a formal PSPACE-completeness proof for a combinatorial game, resolving an open problem from 2017; biologically-informed bone marrow cell classification under extreme class imbalance; and labor economics research presented at a conference in Rome.
The thread connecting these projects is a question I keep returning to: what makes a computational method genuinely useful? Predictive accuracy matters, but so does understanding why a model works. That question led me from building CNN-LSTMs to analyzing transformer attention maps against known physical indicators, and from proving complexity results to exploring quantum extensions of combinatorial games.
This fall I join UMass Amherst's DREAM Lab as a PhD student, co-advised by Professors Alexandra Meliou and Neha Makhija. My doctoral research will sit at the intersection of machine learning and data management, applying learned models to core data systems problems and using database concepts like provenance and causality to make ML pipelines more transparent and trustworthy.
