About
I'm one of the founders of Rowan, where I lead product and strategy. At Rowan, we're accelerating molecular simulation with ML and building tools to help chemists harness those simulations. Before starting Rowan, I attended Northeastern.
Benchmarking NNPs
Neural network potentials are ML models used to approximate solution of the Schrödinger equation, which describes how systems of atoms will behave with quantum accuracy. With help, I maintain NNP Arena, a leaderboard of NNP benchmarks.
I'd also recommend checking out:
- Jin Xiao's list of NNPs,
- Matbench Discovery, a solid-state thermodynamic stability benchmark, and
- MLIP Arena, a collection of benchmarks beyond energy & forces.
Writing
I occasionally post on Substack.