Python, Test-Driven-Development, and AI

AI can write a working function in three seconds, but it can also hallucinate a flawless passing test for a feature that doesn't actually work. This session exposes the hidden "gotchas" of mixing LLMs with the traditional Red-Green-Refactor loop, cutting through marketing hype to address the friction points that software engineers, tech leads, and TDD practitioners face daily. We will dive straight into real-world traps, exploring how silent logical fallacies slip into pytest suites, why automation triggers "lazy refactoring," and how over-reliance on AI can accidentally balloon technical debt.