ML for protein engineering is incredibly fascinating — and pretty much all of it, including RFdiffusion is built on structure prediction models.
This series of talks by Nazim Bouatta is exceptional, helped me appreciate and make sense of these models. Incredible how you can engineer neural nets to learn with way lesser data when you incorporate the right inductive biases: https://youtube.com/playlist?list=PL0NRmB0fnLJQPDZh-6utVnRpF...
I built spoofr as a prank in the last few days - my friends wanted to fake like they were working on something when a teacher at school went by + I wanted to fake my typing speed :p