Pattern fragmentation occurs when features adopt different approaches to the same engineering tasks. AI-generated code can amplify it: plausible micro-decisions vary between generations and developers, even when features work individually. section 4.3.3
Variations can include:
- Error handling with try/catch or callbacks; logging with console.log or a custom logger; validation with manual checks or a library. section 4.3.3
- Across many features, these differences can increase onboarding difficulty, cognitive load, refactoring paralysis, and review friction. section 4.3.3
Figure 4.6 Without defined standards, each AI-generated feature adopts different patterns, creating a maintenance burden and technical debt.
Illustrates how AI-generated features can adopt different patterns without defined standards. section 4.3.3
