Clean Code in AI era
summary
Claims that poor quality code costs about $2.4 trillion a year in the US and that developers spend roughly four days per month fixing bugs lead to concrete fixes: use static analyzers like SonarLint / SonarQube / SonarCloud, enforce quality gates, and apply Clean As You Code to new commits only (about 20% of code changes per year). AI boosts speed ~55% but Copilot’s output is correct ~29% and ChatGPT ~68%, so always review generated code and track cognitive complexity and security issues via telemetry (1.8 billion+ issues seen).
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:06Poor code cost
- 3:56Agenda and tools
- 6:05Pory the poor code monster
- 9:17Poor code attacks
- 10:25Developer struggles
- 12:53AI for developers
- 15:41AI code risks
- 19:44Clean code principles
- 22:50Code issues telemetry
- 26:53Code vulnerabilities
- 29:43Spring Boot best practices
- 31:53Testing pitfalls
- 33:38Code refactoring
- 35:08Java clean code tips
- 37:54Static analyzers
- 41:05Clean as you code
- 43:38Conclusions
related talks
46mCode Quality in the AI era, needed more than ever
49mClean Code in the AI era
43mClean code, is it really worth it? A story about monsters and...
25mClean Code, is it really worth it?
40mCode Rot at Machine Speed: Preventing AI-Induced Technical Debt
19mWith great power comes great responsibility in code quality
14mAI-Powered Code Review and Quality Assurance
33mAccelerate Code Quality in the AI Age