I completed my Bachelor's degree at Harvard, where I studied Applied Math with a focus in signal processing. Through my classes and internships in college I got really interested in topics like decision theory, speech/image recognition, and mathematical machine learning. After graduating I got my Master's at Brown where I had the chance to start digging deeper into the intersection of applied math, statistics, and modern machine learning methods. Since finishing my Master's I've been exploring these areas in the context of biology, first focusing on genomic variant detection and now on drug design. I'm always trying to learn more about ways in which mathematical theory can help us to model biological systems.
I'm a data scientist at Generate:Biomedicines on the Computational Protein Generation team, working on Bayesian methods of uncertainty quantification in various parts of the drug discovery process. While the field of AI-enabled drug discovery has exploded in recent years with in silico methods that have grown to match the scale of available sequence and structure data out there, any workflow that's rooted in experimental data collection will still need to cope with data scarsity. My work focuses building statistical frameworks that can learn across different experiment data sources in order to extract as much information as possible from noisy experimental data sets collected across a variety of conditions and phenotypes.
