Serdias Podcast

. The Migration of Engineering Rigor As AI takes over code production, engineering rigor does not disappear; it shifts to other areas. Rigor is migrating upstream to specification review, as detailed plans and structured formats become essential to produce correct AI-generated code. It is also moving into test-driven development (TDD), which acts as a form of prompt engineering to prevent agents from writing tests that verify broken behavior. Furthermore, rigor is transitioning into risk mapping—matching human verification to the "blast radius" of the code—and leveraging type systems and constraints to make incorrect AI-generated code unrepresentable.2. The Emergence of the "Middle Loop" A new category of supervisory engineering work, termed the "middle loop," is forming between the inner loop of coding and the outer loop of delivery. This involves directing, evaluating, and fixing AI agent output, requiring engineers to shift from direct implementation to delegation and orchestration. This shift creates an identity crisis for developers used to simply translating tickets into code, and indicates an incoming convergence between developer and product management roles.