Theano

Yeah.

I similarly managed a team that we continue to maintain the fork of Theano, which was one of the original symbolic differentiation compilation tools.

Yeah, one of the things that was really interesting about the time at XLA is this is when developers started to really take off with custom kernels for ML development.

And we'd started to see a lot of publications on particular custom kernels or particular techniques for training in particular that... would accelerate the training propagation passes and those kinds of things.