About Kate Borgen

Hi, I'm Kate.

I started coding at 11 and never stopped trying to understand how things work, or how to make them work better. My path into design started in a congressional office, where I spent most of my time manually tracking calls, letters, and emails. Constituent voices got lost in the mechanics of just listening. That shaped how I think about every interface: there's always a real person on the other end, and the tool or system either helps them or it doesn't.

I design and build tools for people who need them most: small business owners learning to use AI safely, citizens trying to reach their representatives, outreach workers serving farmworker communities, cost analysts making sense of defense acquisition data. The common thread is usability with real downstream human impact.

I work across design and engineering as one practice. Some projects I take from concept to shipped product, design and code both. Others I lead a team, mentor apprentices, and stand up a design system alongside the product. I care about scope discipline, research that surprises you, and prompts that read like component specs.

Complex problems need both strategic clarity and technical depth. So I work end-to-end: research and discovery through shipped product, using tools like Claude Code, Lovable, and Figma Make. There's always a sharper interaction, a faster workflow, a clearer path for the person on the other end.

Skills and tools

What I believe

1. Design and engineering are one practice.

   I ship in both. The prompt is a design artifact. The theme file is a design source. The fastest way to lose drift between design and dev is to stop handing things off and start writing them together.

2. Usability has downstream human impact.

   When an outreach worker finds the right tool faster, a farmworker gets better service. When a small business owner understands AI safety, their client data stays private. Every UX decision eventually touches a real person.

3. Scope ruthlessly.

   CivicLink went from 25 routes to 7. AI for Everyman stayed focused on real estate at first. Staff design is knowing what not to build. The hardest decisions are usually about removal, not addition.

4. Trust is designed before the model speaks.

   In regulated and high-stakes domains, the interface has to earn trust before the AI does. Source references, explicit context, grounded data. The pre-AI moment of the interface matters as much as the AI moment.