I design data‑heavy B2B tools, and design systems so people and AI tools build from the same source.
- 10 years designing digital products
- Design experience in B2B and B2C
- I use AI to make my design process faster and more consistent
Open to senior product design roles
Contact meMy work
Projects I've led or contributed to.
Design system for a telecom platform suite
Syniverse's design system was written for people, and an AI agent reads it literally. I extend it and rewrite its docs in Markdown so both can use them. The current version is in production; the next version is in internal QA.
Network monitoring platform
Operators' engineers watch roaming traffic with this tool, and the old one was hard to scan and to personalize. I led its redesign, with one color per protocol built from existing design system tokens. Dashboards and Alarms are in production.
Global pricing platform
A message's price in and its cost out lived in more than twenty tools. I led the design from zero of the internal product that holds both, for the thirteen roles that set, check and pay those prices. Vendor onboarding and gateway control are in production.
Recetas Nestlé
Nestlé needed one recipe platform for many markets. As a UX/UI designer on a team of three, I helped design one base that each market in Latin America configured as its own. It runs today in Mexico, Brazil and across Latin America.
NESCAFÉ® Dolce Gusto México
Home, campaign landing pages and content pages for the brand's online store.
ASSMCA, Puerto Rico
A public mental health app, a volunteer system and an internal incident-reporting tool for a government agency.
How I work with engineering
A screen is not done when it looks right. It is done when it is built from the system.
- Visual QA in staging
- I inspect staging builds with Chrome DevTools and check that computed styles resolve to tokens, not hardcoded values. Gaps become issues: on the monitoring tool, each one went to a joint triage.
- Token compliance is an acceptance criterion
- A screen with a literal color, size or spacing value is not finished. A build step can enforce it.
- Markdown specs so AI builds on-system
- Every component has a Markdown spec. AI tools read it, so what they generate uses the same tokens.
Figma component
Designed once, with its states.
Markdown spec
Written so people and AI read the same source.
Generated UI
Built from the spec, on the tokens.
Verification
Checked in staging against the spec.
This site documents its own design system
Tokens and component specs are written in Markdown, so people and AI tools read the same source. A build step turns the tokens into CSS and fails if a component bypasses them.
The system, as built
Dark themeLight theme
Color steps
Text colors
- Aatext-primary15.9:117.5:1
- Aatext-secondary9.4:19.6:1
- Aatext-tertiary5.7:16.0:1
- Aaaction-text8.9:17.6:1
Type
- Aa
- Aa
- Aa
- 1,024
Spacing
About
I started in brand identity, designing logos and the manuals that keep them consistent. Today I design B2B product systems.
Trained in design and audiovisual communication.
Hiring for a senior product designer?
Send a note about your product and your team.