With Claude: Less Coding, More Testing(henrikwarne.com)
henrikwarne.com
With Claude: Less Coding, More Testing
https://henrikwarne.com/2026/05/31/with-claude-less-coding-more-testing/
5 comments
> These days, I typically start a new feature by asking Claude if the description in the ticket is correct, and if so, I am asking for a suggested solution. I avoid steering Claude to a given solution, even if I have one in mind. Perhaps there are better ways of doing it that I haven’t thought about. I also don’t want Claude to just go along with whatever I suggest.
This is next level dystopian. In your own flow you are not a developer, not even a project manager or a product owne, more like some random consultant thats barely affiliated? Does that make you happy? Do you get any satisfaction out of that flow?
This is next level dystopian. In your own flow you are not a developer, not even a project manager or a product owne, more like some random consultant thats barely affiliated? Does that make you happy? Do you get any satisfaction out of that flow?
I'm not sure there's anything novel about that approach - it's how I've addressed issues quite commonly when assigning them to team members. If you don't prime your preconceived notions on your coworkers(or AI in this case) and let them do their own evaluation from scratch, you can find gaps in yours and their understanding much easier.
What makes you unhappy about this process?
What makes you unhappy about this process?
I found it kinda optimal to "incept" the llm. I better results if I just ask the right questions to guide to to a good solution.
My personal projects have shifted from me writing so much code to me architecting and testing my project more than anything. I also refactor things I don't like code wise to make it much nicer code. Fix small bits of code myself, but you can have Claude write unit tests, and refactor code that still passes the original unit tests.