hire-me@guillermo.design

My design process

I solve business problems. AI just helps me do it faster.

More than 10 years designing enterprise SaaS taught me the hard part was never the pixels. AI made the pixels faster. It didn't make the hard part easier: knowing what to build, why, and for whom.

See how I workView case studies

My design process

The pipeline got shorter.The judgment didn't.

My process remains the same but more efficient given what I can do with the AI technology

  1. Discover

    I talk to users and stakeholders, audit the product, and dig into the data. It's how I make sure we're solving the right problem before anyone designs a thing.

  2. Define

    I turn what I learned into a clear problem statement, priorities, and success metrics, agreed with product and engineering. That way everyone works toward the same goal from day one.

  3. Design

    I explore flows, prototype, and test ideas with real users, building on the design system. Changing a design is cheap; changing shipped code isn't.

  4. Build

    I hand off clear specs and stay close to engineering through QA, adjusting when reality pushes back. That's what makes the shipped product match the design.

  5. Measure

    After launch, I check usage, feedback, and the metrics we set, and feed what I learn into the next cycle. It's how I know the work actually helped.

AI in my process

AI does the legwork.I make the calls.

I use AI every day, as a tool rather than a replacement for judgment. It takes the repetitive work off my plate so I can spend more time on what moves a product: understanding people, making decisions, and getting the details right.

How I see it

Fast isn't the same as right.

AI can produce a hundred screens in an afternoon. Deciding which one solves the problem is still the job.

A draft, not a deliverable.

Everything AI gives me is a starting point that I check, edit, and own.

The system makes it scale.

AI output is only consistent and accessible when a design system sits underneath it, so that's where I invest.

Where it fits in my process

PhaseAI speeds upI own
DiscoverSummarizing interviews, sorting feedback, scanning dataDeciding which problems are worth solving
DefineDrafting problem statements and requirementsPriorities and trade-offs with PM and engineering
DesignExploring variations, writing UI copy, building clickable prototypesChoosing a direction and crafting the details
BuildTurning designs into code and docs with Claude Code and the Figma MCPQuality, edge cases, and review with engineering
MeasurePulling and summarizing usage and feedbackReading what it means and what to do next

What my team gets

More options, earlier.

We compare real directions in days, not weeks.

Prototypes people can click.

PMs and stakeholders react to something working, not a static mockup.

Less lost in handoff.

Specs, components, and docs come closer to the code engineering ships.

More time on research and strategy.

The time saved goes into the work AI can't do.

Let's see if we're a fit.

A 30-minute call is the fastest way to find out. Bring the role, the product, or the problem. I'll bring the questions.

Schedule a 30-min callDownload resume

Prefer email? hire-me@guillermo.design