avatar

Yash Chainani, PhD

machine learning scientist | cheminformatician | computational biologist
Bristol Myers Squibb, San Diego | previously PhD at Northwestern University & Joint BioEnergy Institute, and ML intern at Tatta Bio



Hi! I am a Machine Learning and Cheminformatics Scientist at Bristol Myers Squibb (BMS) in San Diego, CA.

I recently completed my PhD at the Department of Chemical and Biological Engineering at Northwestern University, where I worked at the intersection of machine learning, computational biology, and cheminformatics. I was co-advised by Linda Broadbelt and Keith Tyo, and was a member of the Center for Synthetic Biology at Northwestern and of the new pathway development group at the Joint BioEnergy Institute at Lawrence Berkeley National Laboratory. Prior to my PhD, I earned my BS in Chemical and Biomolecular Engineering at UC Berkeley’s College of Chemistry, where several courses in data science and machine learning sparked my interest in AI4Science.

My research broadly focuses on making ML models more trustworthy and robust in the face of unreliable data - using synthetic data, data augmentation, multimodal contrastive learning and reinforcement learning. Unlike natural language, where large, high-quality datasets are abundant, biology and chemistry often suffer from data scarcity and noise, which makes developing methods that perform well in low-data and/ or imperfect-data regimes especially critical. Most recently, I’ve become particularly interested in leveraging these techniques to build large reasoning models for biology and chemistry. You can find all of my existing publications and select talks on my google scholar page.

News!