Engineers Use LLMs to Hunt Concurrency Bugs Tests Miss

ctojunior's team used LLMs to catch a real concurrency bug that both AI review and the test suite missed. Instead of randomized concurrency testing, they have models deliberately pick and explore thread interleavings, and even used AI to build a testing tool without formal specification languages like TLA+.

2026-09-23 ~ 2026-09-23 · 4 related posts

• Let the Model Pick Thread Interleavings Instead of Random Test Order — cto_junior · 2026-09-23
• Using an LLM to explore thread interleavings and catch a race condition tests missed — cto_junior · 2026-09-23
• Asked AI to Review a Known Concurrency Bug, It Came Back Clean — cto_junior · 2026-09-23
• Dev builds AI-assisted concurrency testing tool that caught real bugs without TLA+ specs — cto_junior · 2026-09-23