Divya Nori
I am a PhD student in Computer Science at Stanford, advised by Prof. Brian Hie at the Laboratory of Evolutionary Design and Prof. Christopher Ré at Hazy Research. I work on AI for biology, and some of the questions I am currently interested in are:
- Along which axes should we scale models to better understand biology?
- Can we unlock emergent properties like in-context learning and reasoning in biological sequence models?
- How should biological sequence models and natural language models interact to make discoveries?
I aim to deploy such models to make advances in health and disease. At the same time, improving biological understanding in AI models raises critical biosecurity concerns. This motivates my work as a founding member of technical staff at Valthos, where we build AI and software infrastructure to rapidly counter emerging biological threats.
Outside of research, I am a co-organizer of the ML Protein Engineering Seminar Series.
Feel free to reach out at dnori [at] stanford.edu
Past
Prior to my PhD, I completed my SB and MEng in EECS at MIT where I became interested in generative biology. My first undergraduate research experience, advised by Prof. Connor Coley, piqued my interest in algorithms for molecular design. I then spent three years at the Broad Institute’s Eric and Wendy Schmidt Center, where I was fortunate to be advised by Profs. Wengong Jin and Caroline Uhler, developing generative methods for RNA and protein design.
Outside of academia, I gained ML engineering experience through internships at D. E. Shaw Research, Absci, and Microsoft Research.
publications
- ICML GenBioBindEnergyCraft: Casting Protein Structure Predictors as Energy-Based Models for Binder DesignThe Forty-second International Conference on Machine Learning (ICML) Generative AI and Biology Workshop [Best Paper], Jul 2025
- ICML ConferenceRNAFlow: RNA Structure & Sequence Design via Inverse Folding-Based Flow MatchingThe Forty-first International Conference on Machine Learning (ICML), Jul 2024
- NeurIPS GenBioEvaluating Zero-Shot Scoring for In Vitro Antibody Binding PredictionThe Thirty-seventh Annual Conference on Neural Information Processing Systems (NeurIPS) Generative AI and Biology Workshop [Spotlight], Dec 2023
- NeurIPS AI4ScienceDe Novo PROTAC Design Using Graph-Based Deep Generative ModelsThe Thirty-sixth Annual Conference on Neural Information Processing Systems (NeurIPS) AI4Science Workshop, Dec 2022