“adversarial review with a fresh context window, running on a different model” is the single highest leverage thing you can add to your flow today that doesn’t ship with any existing harness
austin (top/chicken 🐔)
correct. no code - whether written by a human or an agent - should be getting reviewed by a human without being first reviewed by another agent.
imo that includes reviewing yourself! respect your own time and dont review the agents work until it’s asked for review from another model.
Nuclear opinion: I don’t care if you use AI to write code,¹ but making a human review a change before an LLM is disrespectful.
Why am I catching your bugs if a computer could have done it? It’s like asking to review code that doesn’t compile or pass CI.
¹ As long as you do it responsibly.
