logoalt Hacker News

JeremyNTyesterday at 3:24 PM2 repliesview on HN

I feel like you could have correctly stated this a few months ago, but the way this is "solved" is by multiple agents that babysit each other and review their output - it's unreasonably effective.

You can get extremely good results assuming your spec is actually correct (and you're willing to chew through massive quantities of tokens / wait long enough).


Replies

nicoburnsyesterday at 6:04 PM

> You can get extremely good results assuming your spec is actually correct

Is it ever the case that the spec is entirely correct (and without underspecified parts)? I thought the reason we write code is because it's much easier to express a spec as code than it is to get a similar level of precision in prose.

show 1 reply
ldngyesterday at 3:36 PM

And unreasonably expensive unless you are Big Corp. Die startups, die. Welcome to our Cyberpunk overlords.

show 2 replies