1 min lesson
When given an AI-product bug, classify before you investigate
Put this lesson point into plain words: "Naming the layer first shows the mental model of the architecture."
Step 1 of 2
When given an AI-product bug, classify before you investigate: “This reads like a context-retrieval failure, not generation, so I'd check indexing and the actual @-context payload before I blame the model.” Naming the layer first shows the mental model of the architecture - VS Code fork, client talking to model backends, retrieval feeding context - that the product round is looking for.
Don't reflexively blame the model. Most loud “the AI is dumb” reports root-cause to context or retrieval - the model never saw the file, so of course it ignored it. Blaming the model when the bug is in indexing sends the report to the wrong team and burns a day.