If you think like that why invest in software at all; the AI will do everything?
Does AI make reading or writing stacked PRs any nicer? No, it does not.
If you think like that why invest in software at all; the AI will do everything?
Does AI make reading or writing stacked PRs any nicer? No, it does not.
> If you think like that why invest in software at all; the AI will do everything?
Correct, hence the "SaaSpocalypse" phenomenon in recent weeks. Investors are slowly becoming disinterested in investing in software anymore precisely because models are good enough now to replicate any SaaS pretty easily, which still requires effort but is less so than paying for a SaaS particularly in large organizations which are charged per seat.
It does though.. you don't have agents that can connect to github or wherever your git mirrors are and comment on PRs?
The comments stop me from marking MRs with bad issues as ready, but if reviewing it's not really helpful.
Maybe if I were reviewing some random dude's code, where I have no idea what he's been working on...
Don't you read the PRs?
Yes, but my agents also do.
The whole concept of PR is that you want additional looks on the code, and multiple agents working adversarially on PRs with philosophical rules are really nice.