Question
Greenfield Go service: an AI agent will build a CSV-to-Postgres importer where admins upload customer records (up to ~2M rows), and you need per-row validation, a dry-run preview, and a partial-failure report rather than all-or-nothing. Describe how you'd direct the agent — streaming vs. load-all, the validation strategy, transaction boundaries, and acceptance criteria. Where does a one-shot 'import this CSV' prompt fail at this scale?
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.