News for hackers. Stuff that matters.
Re:In 27,280 coding agent runs, 64% of inference spend went ... by krrrh on Tuesday November 17, 2026 @12:06PM (#50534516)
> it doesn't work like that, for that matter we don't work like that.

The parallels with self-driving cars are interesting. We demand that agents don’t just produce better code than most humans (which the frontier models do), but that they perform much much much better.

There are no comments on this comment yet.