Episode 053: Moving Correctness Out of the Reviewer's Head: AI-DLC v2 vs v1

AWS's AI-Driven Development Life Cycle was rebuilt from the workflow layer up — from v1's tool-agnostic markdown rules with a human approving at every stage, to v2's Skills-based architecture where machine-checkable post-conditions, enforced by a process checker the AI can't modify, decide how much a coding agent runs on its own. This deep dive traces the three structural moves behind the redesign — composition from stages to Skills, correctness from the reviewer's head to explicit contracts, and autonomy as a function of codified verification — then interrogates the claim the specification never proves. The architecture is coherent and unusually candid about its own weak spot, but the headline metrics are vendor-reported, the peer-reviewed research on AI self-verification is unforgiving, and the governance of self-evolving rules and the SKILL.md supply chain remain wide open.