I’d posit that well over half of tokens spent should go not just to write new code/software alone, but in review, testing, as well as tooling. Every bit of code an agent writes should at the very least have an adversarial review loop.
I totally agree, and the differential token cost is minimal if they are included during product code generation.
Anyone who is responsible is doing that.
From the tone of the blog post, it sounded like an untenable problem.
Hard to keep up because there's a lot of work.. that's not untenable, that's just a lot of life.
my rebuttal to his point is that AI generated code does not essentially have to be low quality. And it very well mean higher quality code if supervised properly