Good design starts with understanding what makes a problem difficult. I look at how people work, how information connects and what the team can realistically deliver.

Then I turn that understanding into product structure, clear interactions and working prototypes.
The methods I use depend on the question we need to answer.

I ask people to walk me through real tasks, including the workarounds they have stopped noticing. This helps uncover the rules and dependencies a feature request may leave out.

On DocFlow, research with methodologists, legal, audit and risk teams informed the structure of the document registry. Jobs-to-be-Done helped clarify what people needed to accomplish. Object-Oriented UX helped me map the documents, versions and relationships the interface had to represent.

That structure gave us a shared basis for discussing navigation, workflows and scope.

I make ideas tangible early, then add detail as the questions become more specific. I look at whether people understand their options, what they expect to happen next and where the flow breaks down.

Tour Builder also taught me to revisit assumptions about what information matters. We initially overlooked people who cared about the tour operator itself. Restoring operator names and filtering addressed a choice we had oversimplified.

I carry those findings into the interaction details: hierarchy, wording, defaults, states and recovery paths.

For DocFlow, I built an AI-assisted workflow that connects interview research with custom agents, skills, and our design system. It produces working interfaces and reusable code for engineering handoff.

I review the research findings and design decisions throughout. Working prototypes let me explore complete flows, role differences and interface states, while keeping the underlying research available as the design evolves.

I work with developers as the interface takes shape, resolving questions and reviewing the implementation. After release, I use product data and user feedback to guide further changes.

The measure depends on the problem. For DocFlow’s approval redesign, approval time fell from 14 to 9 days. At Travelata, Tour Builder helped increase conversion by 7.5% and sales by 11.2% over six months.

These outcomes give us a reason to examine what worked and where the next improvement should come from.

My contribution also includes mentoring, design reviews, workshops and reusable tools. I want the reasoning behind a decision to be clear enough for others to question, develop and apply.

At Burning Buttons, I mentored two interns who were later hired full-time. At Travelata, reusable no-code landing-page templates enabled marketing to launch campaigns without developer support.

Helping people work more independently is part of how I approach design.