Research
The self-driving lab and the reproducibility promise
Four papers trace the self-driving lab from a 2019 landmark to today's biology, optimization, and safety work. The through-line is reproducibility, and its unfinished edges.
Research
Peer-reviewed and preprint research on liquid handling and lab automation, dissected and explained for the people who run the deck.
- Research
From protocol text to executable method: what the LLM papers really solve
Three recent papers turn human-written protocols into machine-executable methods. The interesting part is not the language model, it is how they check the output.
- Research
Teaching a machine to pipette: two papers on the act and the choreography
One paper puts a manual pipette in a robot's hand; another treats a full run as a routing problem. Together they show automation is two problems, not one.
- Research
Reading a liquid from how it moves: machine learning at the pinch-off
A 2025 paper shows a droplet's shape as it breaks off carries enough information to predict viscosity and surface tension. Here is what that means for anyone tuning a class.