Generating code is the easy part -- understanding it is hard. As AI accelerates development, it also accelerates complexity, making it hard for humans to keep up. The result? A growing mass of hard-to-maintain legacy code that no one truly understands. (But hey, at least it was fast to get there.)

This talk explores how to reap the benefits of AI without drowning in technical debt. We'll cover the latest AI research together with strategies to enforce human readable code such as automated reviews and the concept of AI guardrails in our IDE. All techniques will be demonstrated on real-world codebases, showing how to keep tools like GitHub, Copilot and Cursor from becoming legacy code accelerators. If you're a software developer or technical leader, this talk is for you. You'll walk away with actionable techniques to manage AI-assisted coding and prevent it from becoming tomorrow's source of technical debt.

‍
Dr. Markus Borg is a senior researcher at the intersection of software engineering and applied artificial intelligence. His research interests include softwareengineering intelligence and AI engineering in the automotive domain. Markus serves on the editorial board of Empirical Software Engineering and is a department editor for IEEE Software.