I’m a product designer who specializes in making complex technical systems intuitive and actionable. Over the past 10+ years, I’ve focused on enterprise platforms, developer tools, and data-heavy applications—designing experiences that help technical and non-technical users alike make better decisions faster.

Currently at GitLab, my focus is on analytics and AI experimentation tools. My work spans from conducting user research to defining data visualization standards used across the product, to prototyping AI-powered data discovery features.

How I work

I’m a product designer who works like a design engineer. I don’t just hand off mockups and wait, I own the data-visualization section of GitLab’s design system as code, and I ship changes to it through merge requests myself. The distance between a design decision and the thing users actually run is where most product quality is lost. I’ve spent my career closing that distance, and now with the advances in AI, the gap is collapsible in a way it wasn’t three years ago.

This changes four things about how I work:

Research synthesis at a different scale. I run research-heavy projects, multi-persona interviews, Jobs-to-be-Done studies, support and usage data. AI lets me theme large transcript sets in a fraction of the time, so the bottleneck moves from processing to judgment.

Divergent exploration without the cost. Dashboard and data-dense work lives or dies on layout, density, and information hierarchy, and the only way to know what works is to see many real options against real data. AI lets me explore and validate multiple directions quickly.

Prototypes that run, not prototypes that pretend. For data and AI features especially, a static, happy-path Figma mockup will always be far from what will be implmented in reality. Designing for non-deterministic systems is a craft, and you can’t practice it on a static frame.

Leverage that compounds across the org, not just my own output. Owning design-system components as code means an improvement I make ships everywhere those components are used. AI accelerates the component work; the multiplier is the system.

Overall, AI multiplies output, but output was never the constraint. Taste, research, and judgment about what is worth building and what ‘good’ looks like are still the bottleneck. AI makes these skills more valuable, not less, now that everyone can now produce volume. My job is to be the person who decides what’s worth multiplying.

Work history

See my LinkedIn for past work experience.