Question
Greenfield: you're directing an AI agent to build a resumable file-upload endpoint in TypeScript (Node/Fastify) backed by S3 multipart, for files up to 5 GB over flaky mobile connections. Describe how you'd scaffold and sequence the build with the agent: how you pick the chunking protocol (tus vs. S3 multipart vs. roll-your-own), what acceptance criteria you hand it up front, and where you insert review gates. What would a careless one-shot prompt get wrong here?
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.
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.