t07: Race condition -- GPT-6.1 Sol vs Claude Opus 5.5
Fix a Node.js job queue that loses, repeats and overruns jobs when hundreds arrive at once.
Opus: 89.0 s. Sol: 486.3 s. Sol took 5.5 times as long.
Opus (s)Sol (s)
Hidden tests: Opus 6 of 6, Sol 5 of 6.
Blind judge pick: Opus (clear).
Sol needed 1 review round. After it: 6 of 6 hidden tests, and the judge said yes.
Sol's re-check after its review round was judged by DeepSeek, not Gemini.
"Solution B [Opus] correctly separates the asynchronous save phase from the synchronous drain loop, maximizing concurrency while preserving order. Solution A [Sol] awaits store operations inside the drain loop, which serializes job preparation and creates subtle re-entrancy issues that can exceed concurrency limits."
"An exceptionally well-crafted solution that perfectly resolves all race conditions using synchronous state management and clever promise chaining."
"The solution implements a robust sequential preparation phase and fixes many edge cases, but fails a hidden concurrency limit test."