“We had a 100 million compound library that didn't help as much as better characterizing a smaller one… you can either spend more on sequencing a smaller set or get more molecules and hope noise averages out”
Latent Space: The AI Engineer Podcast · Lab-in-the-Loop at Compute Speed - Ian Quigley, Leash Bio
Play from 36:19
36:19
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
chose to focus on a smaller, well-characterized compound library instead of a 100 million-molecule library to stay within a fixed budget
100 million compounds library; smallest custom library ~50,000 molecules
Said nearby on the same subject
“We got the chemical library... 130 million molecules... after 2000 teams competed, the winner was a random notebook; no one did better than chance on the most challenging eval”
♫
Said on Latent Space: The AI Engineer Podcast
Growth tactic130 million molecules; 2000 teams; winner was a random notebookShare this moment
“I decided it wasn't worth it… I chose instead to focus on my business… we refreshed our brand.”
“He built more efficient data structures than the standard coding libraries had, ... for the C++ STL map and in Go for the Swift table implementation.”