Tue, Mar 17 · 6:00 PM CDT
Let's be real for a second: if you ask most developers today, they’ll tell you AI coding assistants have made them significantly faster. But if you ask their managers, you often hear a very different story. Engineering leaders are seeing short-term bursts of speed followed by a mountain of rework, architectural drift, and security issues.
The truth is, AI is incredible at producing output, but creating actual business value—and avoiding a tangled mess of "AI slop"—takes a lot more effort than just throwing vague prompts at a model. When we rely on unstructured "vibe coding," AI simply amplifies our ambiguity, churning out inconsistent and unmaintainable code.
Spec-Driven Development (SDD) is how we fix this. SDD shifts the developer’s role from writing syntax to defining precise, machine-readable intent before any code gets generated. In this talk, we'll dive into how treating specifications as executable constraints keeps AI on track, bridging the gap between moving fast and actually building software that lasts. Plus, we'll get hands-on with real .NET workflows to show you exactly how to put this into practice.