Code RoomCode comments match actual behavior
EasyPrep Room Coding #4439

Code comments match actual behavior

Vibe & agenticAlgorithms & data structuresEntry–Mid~12 min

You ask an AI to add explanatory comments to a confusing piece of Bash that nobody on the team understands. It returns this. The comments are confident and readable — but does each one match what the code actually does?

Implement
select_logs_to_compress(file_names: list[str], file_sizes: list[int], max_bytes: int) → list[str]
Examples
in[["app.log","debug.log","notes.txt"],[12000000,500,99999999],10000000]out["app.log"]
in[["*.log"],[5],10000000]out[]
in[["a.log","b.log"],[10000001,10000000],10000000]out["a.log"]
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 12 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.