How Schrödinger sped up molecular discovery by 4x with Alphaevolve
This title could be clearer and more informative.Try out Clickbait Shieldfor free (5 uses left this month).
Schrödinger partnered with Google Cloud to apply AlphaEvolve, Google DeepMind's evolutionary AI coding agent, to optimize their machine-learned force field (MLFF) pipeline. Two bottleneck algorithms — neighbor list computation and Ewald summation — were targeted. AlphaEvolve replaced slow for-loops in the Ewald summation with parallel batch matrix multiplication, raising the program success rate from under 1% to over 60% and boosting the performance metric from 7.9 to nearly 30. The result was a 4x speedup in both MLFF training and inference, compressing molecular screening timelines from months to days across drug discovery, catalyst design, and materials development.
