I've used Claude to write code, and it is much harder to test that code than it is to test code "haphazardly" as I write it myself. Reason being, I can test mine after each new line I write and make sure that line is doing what I intend it to do. After Claude writes a whole set of functions, it could take hours to test all the potential failure modes.

BTW, if it doesn't take you hours to test the failure modes, you're not thinking of enough failure modes.

The time savings in writing it myself has a lot to do with this. Plus I get to understand exactly why each line was written, with comments I wrote, not having to read its comments and determine why it did something and whether changing that will have other ramifications.

If you're doing anything larger than a sample React site, it's worth taking the time to do it yourself.

Well, you could also generate the tests by CC, check them to make sure they’re legitimate, then let it implement it?

The main key in steering Claude this month (YMMV), is basically giving tasks that are localized, can be tested out and not too general. Then you kinda connect the dots in your head. Not always, but you can kinda get gist of what works and what doesn’t.