Dashboard slowness investigation shape
Ticket: "Make the analytics dashboard faster." It's a Python/SQL app where a few queries against a 200M-row Postgres table feel slow. You want to use an AI agent but the ticket has no clear shape. Break it into agent-sized steps with checkpoints, and identify which step is actually a human-led investigation the agent can only assist with, not own.
Implement
rank_query_bottlenecks(stat_lines: list[str], min_share_percent: int) → list[str]Examples
in
[["dashboard_summary|2|1800","user_events_page|900|40","audit_export|1|5000"],20]out["user_events_page"]in
[["slow_but_rare|1|9000","hot_and_small|3000|10"],50]out["hot_and_small"]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 24 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.
Run or narrate your approach, then ask the coach.