I fix the problems nobody's filed a ticket for.
Not broken. Just quietly ignored. That's usually where I start.
Interactive comparison of the same operations dashboard, before and after a redesign. Before: a dated 2010 style portal with cramped tables and plain colored status text. After: a modern dashboard with the identical data, 12,480 active accounts, 94.2 percent retention, 78 percent satisfaction, and the same three regions shown with rounded status pills instead of plain text. Use the slider below, operable by mouse, touch, or the left and right arrow keys, to control how much of each version is visible.
So, about that ticket.
I've been at Zeus Learning since 2021, where six or more products, start to finish, are mine to own directly. I don't get to hide behind a team when something ships broken, and I don't get to share the credit when it works either. Somewhere in there I got good at spotting the stuff everyone else had already learned to work around.
I work best with people who'll just tell me when something's off instead of being polite about it. Lately a good chunk of my time has gone into Weave, an internal tool we're building on Claude, so I've spent nearly as much time thinking about how AI changes design work as I have actually doing design work.
Away from a screen, I'm probably dancing badly, attempting a recipe above my skill level, or hunting down a version of a dish I can't get at home.
Design
Research
Dev fluency
One company, five years, and nowhere near bored yet
- Own UX and UI end to end across six or more live software products, driving each one directly from research through delivery.
- Take interfaces that technically still work and turn them into ones people actually want to use, without breaking whatever's quietly holding the current version together.
- Core contributor on a small team building Weave, an internal AI design tool running on Claude. More on that under Work.
- Defend design decisions directly to PMs, engineers, and leadership, and push back when a request would quietly break usability.
- Run internal user research, surveys and stakeholder interviews included, to decide what's actually worth redesigning and why.
- A 45 day internship getting hands on with MySQL and DBMS basics, and building a low fidelity prototype app for the ground staff department.
Five real projects. Nothing made up.
Tap any card for the full story: role, team, timeline, and exactly what happened, unfinished parts included.
Four steps. Nothing fancier than that.
Discover
Talk to the people actually using the thing, run surveys and stakeholder interviews, and find out what's really broken before touching a pixel.
Define
Decide what's actually worth fixing first. Sometimes accessibility before aesthetics, or structure before polish.
Design
Wireframes, prototypes, and enough rounds of feedback to catch the version that quietly overloads people before it ships.
Deliver
Stay in the room through launch. Present decisions directly to stakeholders, and fix whatever testing turns up.