AI Can Find the Code. It Didn't Know How the System Worked
The article discusses the challenges faced when using AI coding agents to modify complex codebases. Despite the AI's ability to locate files, it often lacks the necessary understanding of how the entire system operates, leading to incorrect modifications. A pilot study revealed that even when provided with detailed information, the AI struggled to implement fixes due to its inability to grasp non-obvious dependencies within the code.
- ▪AI coding agents can find files but often fail to understand the overall system architecture.
- ▪A pilot study using the Claude Sonnet 4.6 AI showed a bug-fixing success rate of 61.9% when given full commit descriptions.
- ▪The AI's failures were attributed to its inability to recognize non-obvious connections and dependencies in the code.
Opening excerpt (first ~120 words) tap to expand
Posted on April 26, 2026 by Adam Wespiser A quick change for a simple feature. The task was to add a basic UI panel to show a warning, a couple API calls to verify references, and a little data validation. The hard part? Modifying a complex monorepo sitting at the core of our business. Not large like “been working on it for a while”, but large like thousands of contributors going back to the second Bush Administration. Even worse, the sections of code I’d be working on laid dormant for nearly 10 years. But this wasn’t the problem, it was something else: the solution depended on information that wasn’t locally visible. The expectation was simple: AI would entirely compress the onboarding and we’d ship something fast. Get in and get out, and let the AI do the work.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Wespiser.
Discussion
0 commentsMore from Wespiser
