ParentFull threadHellDunkel·Is it true that current LLMs can find bugs in complex codebases? I mean, they can also find bugs in otherwise perfectly working codeView on HN