I recently joined Manas AI, an AI-native drug discovery company based in New York and co-founded by Siddhartha Mukherjee, Reid Hoffman, and Ujjwal Singh. We build science-based models to discover and develop new medicines, starting with treatments for cancer.
Before that, I was a founding research engineer at Brightband, where I worked on diffusion models and data assimilation for fully end-to-end AI-based probabilistic weather forecasting.
I was a PhD student at ETH Zürich where I studied machine learning.
Towards the end of my PhD, I interned with Stephan Hoyer for one year, advancing the state-of-the-art of machine learning for computational fluid dynamics, which resulted in this paper published in TMLR. I learned so much from Stephan, Dmitrii Kochkov, and the rest of the team.
I wrote my dissertation and found my next position working at AI2, developing pure ML-based climate models under the leadership of Chris Bretherton. I contributed to the groundbreaking first pure AI-based climate model.
My research interests include convex optimization, Bayesian inference (variational inference), simulation, inverse problems, probabilistic programming, and more. Over the past few years I have focused on ML for weather and climate, and more recently on ML for drug discovery.
Outside of work, you can find me playing a Pete Ross banjo.
Last updated: June 10, 2026