Write to think first, then prompt. Pressure-test the problem before you ask AI for solutions.

Before I open any AI tool, I just dump everything I know. PRD stuff, research, customer insights, what I think is broken, half-baked ideas.

It usually makes the gaps pretty obvious.

Only after that I open Claude. And not to design anything yet—just to poke holes in my thinking:

• What am I missing?
• What assumptions am I treating as facts?
• Where does the PRD not match the research?

At that point, it's more about pressure-testing the problem than jumping to solutions.