Ryan Keisler

I'm a versatile physical scientist who enjoys working with fast-moving teams on challenging and important problems.

I've had the opportunity to work across diverse domains, ranging from cosmology to remote sensing, from mineral exploration to weather forecasting. A consistent focus has been deriving insights from sensor data by combining physical principles with statistical and machine learning approaches.

In 2024 I jumped back into the world of AI weather forecasting, as cofounder and Chief Scientist of Brightband.

Previously I was a Staff Data Scientist at KoBold Metals, working at the intersection of sensor data and data science to discover new deposits of battery metals.

In 2021 I independently conceived and executed on a high-impact research project for ML-based weather forecasting, Forecasting Global Weather with Graph Neural Networks (paper, site, thread, press).

I was previously the Chief Scientist at Descartes Labs, a geospatial software and analytics company.

Before that I was a cosmologist at the University of Chicago and Stanford University, focused on the cosmic microwave background and large-scale structure. I pursued most of this research working on the South Pole Telescope, a project that took me five times to the South Pole in Antarctica - an incredible place!

Experience

Education

Selected projects

Selected publications from my cosmology days

email
linkedin
github
twitter