That's a bummer! At my company we've started investing in what I'm calling 'semantic linting', which is basically running GPT over a PR with a set of rules that we iterate on. Already I'm finding huge value for style/pattern comments that linters can't easily catch, dropping warnings for common DB migration footguns, or notifying people of changing patterns/new ways of doing things. Been great so far!
Do you have any write-up about this or more info? It sounds like a useful use case but I haven't yet got it right
I do not, but it took me and a coworker all of an hour to setup. Create a CI workflow, vibe code it to load up the PR contents + any rules you want (we maintain a directory of individual rules that people can tweak), and ship it off to GPT for a response. Add some input/output schemas, and it's pluggable back into the CI hooks to notify of failed build steps (Github in my case).