Uxia

Uxia is a user testing platform where the participants are synthetic. A team sets up a study — the flow to test, who to test it with, what they are trying to find out — the testers run it, and the findings come back the same afternoon instead of two weeks later.

I joined as founding designer, so there was nothing to inherit — no product experience, no visual language, no components — and what needed designing was unfamiliar in itself: synthetic participants, model-driven runs, a finding that did not come from a human being. Researchers are trained to be sceptical of their data, and rightly so. So the work was two things at once: building the foundations, and turning genuinely complicated concepts into something a researcher can act on without a manual.

I designed and shipped features end to end rather than handing off screens. That meant framing the problem before drawing anything, prototyping to find out where a flow actually breaks, and staying with it through implementation — reviewing the build, going into the code where that was faster than writing a ticket, and closing the gap between the design and what went live. All of it came back to one principle: making the testers understandable rather than magic, showing the run rather than only the summary, and tying every finding back to something you can go and look at.

Beyond the product I led the design of the company website and the marketing around it. A startup site is the first thing most people meet, so it has to say the same thing the product does: same language, same tone, same promises. Aligning brand, product and business goals in practice meant the marketing did not over-promise what the product delivers, and the product did not feel like a different company once you signed up.

The images above are placeholders while the product screens are prepared.