Have you seen early productivity gains from AI, only to watch them disappear under growing complexity and production bugs? You're not alone. There's a common reason: many production systems already struggle with technical debt. When AI agents enter the development loop, that debt becomes a multiplier. Poor-quality code not only increases defects and costs. It dramatically raises AI risk by driving high breakage rates, turning promising agents into legacy code generators rather than genuine help.

Fortunately, there's hope on the horizon. In this masterclass, Adam Tornhill shows how technical debt management now determines the success or failure of AI-assisted development. Drawing on Your Code as a Crime Scene, behavioral psychology, and large-scale empirical studies on AI coding, the course explains AI-friendly code and demonstrates how low-quality systems make AI fail.

You will learn how to assess AI readiness using behavioral code analysis, identify high-risk hotspots, quantify the costs of technical debt, and prioritize remediation efforts. The course also demonstrates how to safeguard code and guide AI-assisted refactoring so both humans and machines can reason about and evolve the system. As a bonus, you'll learn what code that works well with AI looks like and how to avoid the cognitive traps of AI-assisted coding.

This is not a course about tools or prompts. It is about understanding the cognitive and technical forces that shape software systems in an era where machines have joined your development team. You will leave with hands-on techniques to benefit from AI coding agents, achieving speed with quality rather than trading one for the other.

Prerequisites

The masterclass is programming-language neutral. No code will be written, but participants should be comfortable reading and discussing source code.

Target audience

Senior developers, engineering managers, and technical leaders adopting AI-assisted development.

Style

Hands-on in front of your laptop. The masterclass is based on the book Your Code As A Crime Scene (2nd ed, 2024) and empirical benchmarking studies such as "Code for Machines, Not Just Humans: Quantifying AI-Friendliness with Code Health Metrics". The focus is on research and evidence, not opinion. During the class, we'll also use the CodeScene tool to automate the behavioral analyses.