Code RoomDesign doc overstatement review
HardPrep Room Coding #4434

Design doc overstatement review

Vibe & agenticAlgorithms & data structuresSenior–Staff~22 min

You used an AI to draft a design doc for a new event-driven order-processing pipeline based on your bullet points and a few code sketches. The draft reads beautifully and confidently asserts the design 'guarantees exactly-once processing,' 'scales linearly to millions of events,' and 'has no single point of failure.' How do you review an AI-written design doc for overstatement before it goes to a design review, and what's the cost of letting these claims through?

Implement
flag_overstated_claims(doc_lines: list[str]) → list[str]
Examples
in[["The pipeline guarantees exactly-once processing.|","Latency budget is 200ms per event.|see load test 2026-03"]]out["The pipeline guarantees exactly-once processing.","missing: limitations section"]
in[["Non-goals: cross-region failover is out of scope.|","Throughput scales linearly with partitions.|benchmark: 4 brokers, 12k events per second"]]out[]
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 22 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.