I don’t understand this take. In my company a lot of the Claude Coders seem to be very uninterested or unaware of the code they are producing, while I in Cursor usually click ”Keep”/”Undo” on specific code blocks (with little edits) or sometimes the whole file at once if it’s a low risk part of the codebase. I fail to see how this workflow produces inferior code vs shooting in the blind and maybe skimming a huge diff in one go.

I use Cursor and it’s been fine. I write a lot of code manually too, so I liked the tight integration with VSCode, my daily driver for about a decade. I used to use Vim, so I’ve “discovered the terminal” a long time ago.

The people steering the agents are the ones producing low quality code. I see little correlation outside of that.

I’m running Opus through Cursor like I would in Claude code, so make sure you’re not comparing different models.

But the core difference for me is I can easily see the code changing in cursor, but not in Claude code. Bigger tasks, I’ll have to drop into other tools to see what Claude is doing along the way. I don’t like that. I like everything being in the same spot