Claude confidently pointed me to a code path in MongoDB’s massive infrastructure codebase. I spent an hour investigating it before realizing something was off. The path worked, technically. But there were better approaches - ones the agent had completely missed.

The fundamental challenge with using AI in large, established codebases isn’t about the AI’s coding ability—it’s about the AI’s lack of context. In greenfield projects, AI thrives because there’s little existing context to understand. But in enterprise codebases with millions of lines of code, the AI is flying blind unless we explicitly give it the right information.

This is the paradox: AI has近乎无限的 intelligence but有限的 information. And intelligence without information is squandered.

Full article content from benr.build