AI has changed how I work, not how I think
I use AI as a real part of my process, not just to rewrite emails.
It helps me move faster, explore more, and spend more time on the work that requires intention and focus.
Accelerating ideas to working prototypes
Using AI to build fully functional prototypes has reignited a giddy excitement I haven't felt since I was teaching myself HTML and CSS by making custom MySpace layouts on the family PC.
The most valuable thing I've learned about AI-assisted prototyping is that the real work happens before you write a single line of code. I spend the majority of my time planning with the AI, working through the spec, pressure-testing the logic, and getting alignment on exactly what we're building before anything gets built.
Once the spec is solid, I use tools like GitHub Copilot, Cursor, or Lovable to translate it into a working prototype at a pace that just wasn't possible before. The result isn't a polished click-through, it's something real that people can actually use, which changes the conversation entirely.
IN PRACTICE
On a recent RFP response, I used GitHub Copilot to build a fully functional contractor mobile app (15+ screens, real interactions, multi-language support) in a matter of days.
Research has a lot of necessary but time-consuming steps: drafting interview guides, transcribing sessions, writing summaries. AI handles the first pass on all of those, which means I can spend my energy on interpretation rather than documentation.
Where it gets really interesting is using AI agents built on the actual research transcripts as a sounding board. I can ask things like "would users understand this interaction?" or "does this label match the language people actually used?" and get a response grounded in what real participants said.
Accelerating research grunt work
IN PRACTICE
On a contractor digital experience project, I built an agent on top of the research transcripts to pressure-test UX decisions.
I could ask things like: "Would contractors understand their progress is being saved if the button just says 'Continue'?" and get an answer rooted in what people actually told us.
Some of the most important design work is also the most repetitive: accessibility checks, responsive edge cases, component variations. AI handles the first pass on all of it, surfacing issues and generating options for me to evaluate and refine.
Leveraging AI to get the ball rolling on these activities ensures they get done, even with the tightest deadlines. These details matter too much to get squeezed out by timeline pressure.
Streamlining workflows without sacrificing quality
IN PRACTICE
On an AI platform project, I used AI to run WCAG accessibility checks on designs, flag responsive breakpoints for edge cases, and generate component variations for review, work that would have eaten hours of a sprint.
What AI doesn’t replace
I use AI to do more and move faster, but never to think less. It accelerates execution and expands what's possible, but it doesn’t replace:
Deep empathy for user needs, developed through research and observation
Strategic judgment about which problems are worth solving
Cross-functional collaboration and stakeholder alignment
Understanding operational realities and organizational constraints