I have been keeping a running list of papers and ideas that I want to revisit later. Until now, that list has mostly lived in private notes and half-finished whiteboard sketches.
This is my attempt to move a slice of that into public writing. I am going to post short write-ups of papers I read and concepts I have been thinking about recently. There is no fixed schedule. Some posts may come out back to back, and some topics may sit for a while before I get around to them.
This is mostly for me and maybe for people in my lab. Still, I am going to try to make these posts usable. That means adding figures where it helps, and writing down the parts that tend to fade from memory first. If any of this is interesting to someone else, or if you have been thinking about something similar, I would love to hear from you.
The first technical post in this series is Data and foundational MLIPs, focused on the role of data in foundational machine learning interatomic potentials.