AI can read and synthesize way faster than I can, so I use it for speed. But the calls about what matters, what’s worth building, and how it should feel stay with me. Here’s how it fits into how I actually work.
How I work, start to finish
I bring people in when it makes sense, not on a schedule.
It’s not a straight line. Research can send me back to the problem, refining can reopen the prototype, and what we learn after launch starts the next loop.
Frame the problem
Before I pull anyone else in, I talk it through with Claude. It pokes holes in the problem statement, surfaces edge cases I missed, and calls out my assumptions, so when the team gets together we’re focused on the right problem.
Where I keep it human
Anyone can get AI to make something. Knowing if it’s actually right for the people using it is the hard part, and that’s still a human job.
AI helps with
I decide
Synthesizing research
Which insights matter
Building prototypes fast
Taste and trade-offs
Reading the data
Empathy for the people using it