EA Hackathon Projects
Almost all projects started at hackathons die young, whether because the team doesn't prioritize continuing work, or because the idea wasn't actually what users needed. So: if you want impact, I think you should contribute to existing projects such as the open-source issues below. Contributing to open-source is also a great way to develop software engineering skills, which are a key bottleneck for AI Safety research among other fields, and I think a day modelled on PyCon sprints can help.
I've selected the issues below to be feasible first-time contributions to open source for people who are already comfortable programming in Python. Even if they seem small, they will take much longer than you expect while you learn the workflows and tools used in serious open-source projects - that learning is part of the point! Please confirm with Zac before you start working on an issue, to avoid duplicate work and ensure it's a good fit for your skills. If you're after a more challenging or specific kind of issue, just ask - this is not an exhaustive list!
- (webdev) Zac-HD/shed#49 webdev to get an online demo of code refactoring
- (feature) Zac-HD/shed#22 new refactorings in
shed- like a toy compilers problem - (feature)
flake8-bugbearhasgood first issues to add automated warnings about suspicious code - (speedup) unionai-oss/pandera#404 refactor dataframe generation for speed
- (internal) HypothesisWorks/hypothesis#3074 add type annotations to Hypothesis
- (website) HypothesisWorks/hypothesis#2943 update Hypothesis' website build system
- (feature) HypothesisWorks/hypothesis#2414 improved generation of random timezones
- (feature) pytest-dev/pytest#10478 warn about impossible-to-satisfy regex filters
- (design) I have several small projects in need of logos, whether drawn by a human artist or generated by AI - Zac-HD/hypothesmith#7, Zac-HD/shed#3
- (governance) Help write up one-page briefs explaining how governments and companies can fund public-good open source projects, and for projects to get shovel-ready to receive funding! I have several ideas here which I think are high-impact and bottlenecked on my spoons for this kind of writing.