If models keep improving, and we eventually inject these code review bots with more and more context to validate even things like business logic, tribal knowledge, etc. do you think we could eventually arrive very soon in a place where we no longer do traditional code reviews?
Just work with your coding agent to develop things, get it reviewed by a code review bot, fix any issues caught, and ship.
(i'm the lead engineer on diamond)
"Code review" as defined as a human in the loop getting the final say on whether a change will be made to the system will be the absolute last thing to go. That process may look very different as both the inputs massively increase in scale and the methods get disrupted by AI.
I know this is probably not something you can divulge but I look forward to how your team at Graphite plans to solve this! (Would also love your personal take!)
We're adding an agentic chat sidebar to the pull request that can help you gather context to understand a PR and make pre-merge changes without needing to leave the PR page. Open beta next week.
We're only going to spend more and more time looking at diffs as the amount of them increases, so the tools that we use here need to evolve as well. "AI code review" should be an interactive experience that directly builds on how code is reviewed today, not just a bot that comments on your PRs.
I guess that they are moderately useful for finding copy-paste errors.
throwmeaway222•6mo ago
palata•6mo ago
Depends on how good your code is :-).