Code RoomHand off gameplay optimization mid-task
HardPrep Room Coding #4385

Hand off gameplay optimization mid-task

Vibe & agenticAlgorithms & data structuresSenior–Staff~22 min

You're mid-task with an AI agent optimizing a hot gameplay loop in a C++ game engine — you've gotten the frame time down but you're handing off to another engineer because it's blocked on a profiling question. Write the handoff so they can continue the AI-assisted work without regressing what you've done. The codebase has subtle constraints: certain systems must run in a fixed order, and some 'obvious' optimizations break determinism needed for the replay system.

Implement
find_order_violations(proposed_order: list[str], constraint_lines: list[str]) → list[str]
Examples
in[["input","physics","collision","render"],["physics|collision","input|render"]]out[]
in[["input","collision","physics","render"],["physics|collision","input|render"]]out["physics|collision"]
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.