Yes, but compared to seeing 3k lines of code in a PR with no history with vibes in the description, what is better? This definitely "feels" like something worth exploring.
Yes, but compared to seeing 3k lines of code in a PR with no history with vibes in the description, what is better? This definitely "feels" like something worth exploring.
We never accepted 3k lines of code in a PR from a human, why do we from LLMs?
And if we're going to, why are LLMs not doing the PR review?
And if we don't trust them, then isn't that a rather large obstacle to the idea that LLMs are going to replace humans in software engineering?
You can defer review entirely to an LLM, only to human reviewers, or both. We are experimenting with both and have configured LLM skills for pull request reviews. On large projects we also configure LLMs to produce commit segregated diffs within a single pull request, to aid human review. This process is using the github pull request workflow that Zed is attempting to replace.
Why is "feels" in quotes?
Anyway, I'm probably being over cynical and yes it is worth exploring alternatives to the current system.
However I already feel overwhelmed dealing with agent chats. I can't currently see how this deals with the pr problem: If I am supposed to delegate to agents why do my conversations need reviewed by my peers. This feels like classic bad management practices of not trusting the staff below you to get on with it, instead turn up in every meeting.
And end up with 3k lines of code + conversation history?