Everyone talks about AI removing bottlenecks. I think that's one of the most dangerous ideas the industry has embraced. Here's what we should optimize for instead.

Should engineering leaders read code or stay focused on strategy? The answer isn't a compromise--it's an entirely different way of thinking.

A first-principles critique of the AI Code Review Bench methodology, arguing that AI code review has split into two distinct problems — human comprehension and machine verification — that no single benchmark can yet capture.

The AI race focuses on intelligence. History suggests we should focus on organization.