Ok, but my problem is that what it builds is not right. It's never right on the first try, or the second or third tries. There's always a ton of incremental corrective feedback needed.
Do you run into this problem? If not, then I need to figure out how to stop running into it myself.
If so, then how do you give it all the incremental corrective feedback without a conversation?
A combination of guardrails - the spec stage is required to give me a mockup for UI work, it has to make failing tests for bug fixes, etc. - and adversarial reviews ensuring the lazy fuckers don't just give up half way through.
20-30 agents probably touch an average ticket, but I sign off about twice.
Do you run into this problem? If not, then I need to figure out how to stop running into it myself.
If so, then how do you give it all the incremental corrective feedback without a conversation?