There's a thing that happens when you've been heads-down on a build for too long.
You've been prompting for hours. Folders are set up, context is loaded, the model is doing exactly what you asked. And yet you hit a wall. Not a technical wall. A you-shaped wall. You just can't see it clearly anymore.
Then you take a shower, go for a walk, or drive somewhere, and the whole thing restructures in your head. The thing you were trying to force out of 40 prompts appears in about ten seconds without any effort at all.
There's actual science behind this. A meta-analysis of incubation research (https://gwern.net/doc/psychology/writing/2009-sio.pdf) confirmed that stepping away from a problem genuinely improves problem-solving. It's called incubation, and it's a recognised stage in the creative process that you can't shortcut. A more recent study in Scientific Reports (https://www.nature.com/articles/s41598-025-09736-y) found that greater mind wandering during those breaks predicted greater creative improvement specifically for the same problem being worked on. Not just rest, actual mind wandering. Neuroscience research more broadly points to the Default Mode Network as the reason: it activates when you stop actively trying, and it's where your brain makes the wider connections that focused, sequential thinking tends to block.
What I find interesting about building with AI specifically: Claude handles so much of the execution load now. But the strategic insight moment, the "what if we approached this completely differently" thought, that still seems to happen in the same places it always did, and sometimes that's away from the screen.
So I'm curious where yours hit.
Drop a comment if you want to share what you were stuck on when the idea finally clicked.
For me, it's always during a shower or a workout 🚿🏋️