William Zeng

Hi! I’m one of the founders of Sweep, a github app that solves issues(like small bugs) by writing pull requests. This looks like a good issue for Sweep https://github.com/sweepai/sweep to...

Hi! I’m one of the founders of Sweep, a github app that solves issues by writing pull requests. This looks like a good issue for Sweep https://github.com/sweepai/sweep to try. We...

@nate-wandb I'm one of the founders of Sweep, an AI powered junior developer. I wrote and reviewed an initial PR using Sweep at https://github.com/wwzeng1/wandb/pull/3/files. I also opened it against main...

Thank you Nate! I suspect it's around 60%+ of the way there, wanted to introduce the tool to you. Please reach out to me if there are issues.

We're building something for this: https://github.com/sweepai/sweep

@AntonOsika I'm taking a look at this

I'm a bit confused, this repository doesn't seem to use npx. I will try prompting Sweep to run the benchmarks on subprocesses with separate ports.

https://github.com/AntonOsika/gpt-engineer/pull/413 tested this but I'm not sure if it's what you wanted. Let me know if you have more feedback!

Sorry to hear that, it's a lot worse in repositories which use LLMs, mostly due to the self referential nature of prompts, etc.

Hope you are able to get value in the future! We're working on improving it :)