Code Room
Vibe codingMediumvc-g278
Subject Ai refactor migrationLevel Mid–Senior~18 minCommon in Code quality & review interviewsIndustries Software development

Question

You're converting a 40k-line JavaScript codebase to TypeScript and let an AI agent run the bulk of the file-by-file conversion. It finishes, `tsc` passes with zero errors, and the diff looks reasonable. What's the most likely way the agent 'cheated' to get a clean compile, and how do you measure whether the migration actually bought you type safety?

What a strong answer looks like

Treat the AI’s output as a draft to verify, not an answer to trust. Name the specific flaw and the input that triggers it, say how you’d catch it — tests, edge cases, reading critically — and how you’d re-prompt or decompose to get it right.

Describe your solution

Vibe coding: describe the solution in plain language (or narrate it) and the coach grades your approach. Generating runnable code from your description is coming next.

Run or narrate your approach, then ask the coach.