I tried this five years ago back when I was an engineer on the PyTorch project, and it didn't work well enough to be worth it. Has it improved since then?

It works well enough that I didn’t realize this wasn’t first party till right now.

It works, but there are fair amount of caveats, especially for someone working on things like Pytorch, the runtime is close but not the same, and its support of certain architectures etc can create annoying bugs.

For me, no. Spend days trying to get it to recreate a production environment workflow. It is too different than production.

it has. it's improved to work with ~ 75% of steps . fast enough to worth trying before push

I tried this recently and it seems like you have to make a lot of decisions to support Act. It in no way "just works", but instead requires writing actions knowing that they'll run on Act.

I tried act, on the surface it seems like a godsend. Not until you try to use it do you realize it's almost impossible to recreate any moderately complex workflow.

You HAVE to run it against a container, so if you're using self hosted runners your environment may not match at all.