Code RoomConfident wrong advice in expertise
HardPrep Room Coding #4362

Confident wrong advice in expertise

Vibe & agenticAlgorithms & data structuresSenior–Staff~18 min

You're a database engineer. You ask an AI assistant how to speed up a slow Postgres query, and it confidently recommends adding `SET enable_seqscan = off` in your application's session and assures you it 'forces the planner to always use the faster index.' It sounds authoritative and even cites how the planner works. Because this is your area of expertise, what do you do with this advice, and what does this moment teach about working with AI in a domain you know deeply?

Implement
review_query_tuning_plan(steps: list[str]) → str
Examples
in[["set_enable_seqscan_off"]]out"reject_planner_override"
in[["explain_analyze","check_row_estimates","add_composite_index"]]out"accept"
in[["add_composite_index","explain_analyze"]]out"diagnose_first"
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.

0:00 of about 18 min

Vibe & agentic: describe the solution in plain language (or narrate it) and the coach grades your approach.

Which questions mattered is sealed until you submit. Telling you now would just be handing over the edge cases.