← All stories
● Covered by 1 source · 1 reportMedium impact1 neutral

Robert C. Martin discusses reviewing AI-generated code with high-level diagrams

🔄 Updated 9h ago
New to BrevFeed? We gather this story from every outlet covering it into one summary — ranked by real-world impact, not just the latest headline — so you never miss what matters. What is BrevFeed? →

Key points

  • Robert C. Martin advocates reviewing AI-generated code via high-level diagrams.
  • He uses a tool that visualizes project structure and identifies code quality issues.
  • The tool highlights problems like high CRAP scores and low mutation testing.
  • This method allows architectural review before code is finalized.

Shift in Code Review Paradigm

Robert C. Martin, known for his book "Clean Code," observes that the speed of AI code generation makes traditional line-by-line code review impractical for humans. He suggests that developers will need to step back from detailed code inspection and focus on higher-level supervision of AI agents.

Martin states he reviews only 5% of the code his agents write, indicating a significant departure from conventional development practices where human readability of code is paramount.

High-Level Review Tools

Martin utilizes a tool that breaks down his project into a diagrammatic representation, similar to UML. This tool allows him to expand, collapse, and drill into specific parts of the project structure. This visual approach enables him to understand the agent's output without direct code inspection.

The tool dynamically generates these diagrams after the code is produced, contrasting with the traditional static UML approach used before code writing.

Identifying Code Quality Issues

The review tool highlights potential problems within the project structure. It flags functions with high CRAP scores (Change Risk Anti-Patterns), which combine cyclomatic complexity with test coverage, indicating complex and poorly tested code. It also identifies low mutation-testing scores and incorrect dependency directions.

This allows Martin to assess the partitioning, dependency directions, and decoupling of frequently changing components from stable ones, ensuring architectural integrity.

Interactive Architectural Review

The agent is integrated with the diagram, enabling Martin to run "what-if" scenarios and request component movements between layers. This interactive capability allows him to review and approve proposed structural changes before they are converted into source code. While he can still access module-level details, he typically does not.

✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →

The daily brief

One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.

One email a day. Unsubscribe in one click, any time.

Today's brief

Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.

~4 min · 3 stories · Nov 21

▶ Play today's brief Listen on Spotify

New every morning, and the back catalogue is archived by date.

Reporting from

Robert C. Martin, author of "Clean Code," suggests that developers will shift from reading AI-generated code line-by-line to reviewing high-level diagrams. He uses a tool that visualizes project structure, identifies issues like high CRAP scores or low mutation testing, and allows for architectural changes before code generation. This approach addresses the speed at which AI agents produce code, making traditional code reviews impractical.