TreeHouse VS MonkeyType

I've done some limited experimentation with this[2], and it seems like a quite powerful technique for programming. I was inspired by this demo[1] of Dion from 2020. (The first video is amazing)

[1] https://media.handmade-seattle.com/dion-systems/

[2] https://github.com/mikewarot/TreeHouse

i can't, unfortunately, it was youtube internal code. but by the same token a lot of the performance stuff was specific optimisations for that code and would not really generalise.

one of the current "extremely large python codebase" projects is instagram, and they do have some public repos, notably monkeytype (https://github.com/Instagram/MonkeyType) which youtube did have its own analogue of, and cinder (https://engineering.fb.com/2022/05/02/open-source/cinder-jit...). in general facebook's engineering blog is a great place to read about this sort of thing.

MonkeyType collects runtime types of function arguments and return values, and can automatically generate stub files or add type annotations directly to your Python code based on the types collected at runtime.

To alleviate the burden of manual annotation, MonkeyType offers a clever solution. It dynamically observes the types entering and leaving functions during code execution. Based on this observation, it generates a preliminary draft of type annotations. This significantly reduces the effort needed to add type hints to legacy code.

https://github.com/instagram/MonkeyType can perform the call logging, and can export a static typing file which is used by mypy, but also e.g. PyCharm. It doesn't expose such fine grained types, but you could build that based on the logged data.

Which introduces MonkeyType, a python library that generatics static type annotations by collecting runtime types.

Not sure if you're joking but there is something similar for python developed by a rather well known company https://github.com/Instagram/MonkeyType