[shusaku]

Claude is absolutely terrible at writing unit tests, this advice continues to perplex me. Coding agents usually slam you with hundreds of lines that do little more than “test” what is obvious. Maybe it’s good for an API revision or such. But for linear algebra where errors might not manifest unless suitably chosen inputs are generated based on expert knowledge, that’ll be useless.

[waynecochran]

Have you used Claude lately? It definitely will outperform a human at this task now.

[shusaku]

I use it literally every day. But please do not use it to generate tests on its own, it’s all sleight of hand. It does not (without substantial guidance and tweaking from an expert) have the capability to generate meaningful tests.