How do lab-in-the-loop ML systems learn most efficiently? How can you pack as much information as possible into an experiment? We explored this in our new preprint “Information-Dense Synthesis for Molecular Discovery” (arxiv.org/abs/2610.08495).

Show Post

We can shape the model by experimental feedback, as we train in a continuous parameter space and use efficient techniques for fine-tuning and reinforcement learning.

Show Post

This is a really cool way to explore sequence space 🧬! For the Cytochrome design, what is the divergence from equal nucleotide probability at each step in the sequence? Is cVS picking a constrained set of designs and making them, or is the structure space large enough to accommodate many sequences?

Show Post

This finely crafted basalt staircase was built, as far as I can tell, for sheep—it lets them cross this lava flow to the grass above

Show Post

Field work in the Faroe Islands—why aren’t more movies shot here? Definitely feels like another planet

Show Post