Notes from the workbench

What we are learning about adapting biofoundation models to real problems: what the models do well, where they fall down, and what it takes to make a result reproduce in a lab.