Define the problem space by partnering early with Product Managers and Engineering leads — everyone stays in their own lane, but alignment happens up front. AI helps explore the opportunity space faster, surfacing more possible directions before committing to one, so the team spends less time mapping and more time deciding.
Capture user insight through interviews and structured research, then let AI help synthesize it — consolidating themes across sessions, organizing feedback logically, and surfacing behavioral patterns faster than manual affinity mapping alone. The goal is still empathy; AI just gets the team there quicker.
Sequence the flow before committing to screens — but today the storyboard isn't the deliverable, it's the bridge. Each frame is scoped to feed directly into a working prototype, so the story gets tested with real customers instead of just presented to them.
This is where AI has changed the process the most. What used to take weeks of build time now becomes a working prototype in days — design, product, and engineering each stay in their own lane, but collaborate tightly to get something real in front of customers fast, iterate on their reaction, and validate direction before any heavy investment.
Implement the design system — typography, grid, color, and reusable components — with AI doing more of the production work. I've used Claude Code to build internal skills that streamline component building, giving engineering a 1:1 comparison between the design and the code so implementation matches intent from the start.
I call this UX Real-Time Testing — validating with customers continuously instead of waiting for a finished build. AI consolidates feedback as it comes in, so qualitative and quantitative signals stop living in separate silos and the gap between them keeps shrinking. In the end, it's not about the artifact — it's about proving customer value, fast.
