> "It’s really easy to write terrible code. But it’s also really easy to write code that follows absolutely every best practice, with 100% test coverage, and has been fuzz-tested and mutation-tested for good measure – your startup will just run out of money before you finish. So a lot of programming is figuring out the balance."
The situation appears to be changing rapidly. 15 years ago there were no advanced code-generating LLMs, and while relying on them for the core logic without having a good understanding of the language and system is still a bit iffy, it does seem that they're pretty good for generating tests and spotting deviations from whatever your group has decided are best practices.
What sort of tests does it seem good at generating? I was hoping it would give me a decent start writing tests with Jest and React Testing Library for an already existing but untested codebase. It seemed ok at small components that I can also quickly write tests for, but the larger, messier things would receive useless, failing tests.
The situation appears to be changing rapidly. 15 years ago there were no advanced code-generating LLMs, and while relying on them for the core logic without having a good understanding of the language and system is still a bit iffy, it does seem that they're pretty good for generating tests and spotting deviations from whatever your group has decided are best practices.