I'm Carlos. I design products where getting it wrong costs something: a missed regulation, a patient dropping out of therapy, a booking flow used by 152M passengers.
I like the dense, unglamorous problems: AI features that have to earn their place, design systems nobody wants to fund, and data-heavy tools built to scale.
1 Product-first: name the real problem, make the tradeoff out loud, measure what changed.
2 AI where it removes work, not where it adds a demo. Prototype it, test it with users, keep what holds up.
3 Systems so the next team builds on shared parts instead of starting from scratch.
4 Built to scale: it still makes sense at a million users, and to the team maintaining it.
Work
View all
DatamaranDatamaran design harnessA design harness that turns Datamaran's design judgment into rules AI agents and engineers can read: used to build production screens, and as the environment every prototype runs in.
DatamaranWhen the AI says "I don't know"Redesigning how an LLM's "unknown" becomes a short question the user can answer, with the legal source one click away.
DatamaranDatamaran architecture redesignRestructuring how a complex ESG platform reveals what matters — reorganizing navigation around jobs, not features.Writing
View allBuilding
View all- Product buildingHuecastA 2016 side project rebuilt ten years later. The idea held: colour carries the forecast, the numbers come second.
- SystemGlyph SystemThree shapes, five relations, one glyph per post. I designed the grammar and a reference graph; AI turned them into a system that maps each post's idea to a shape, and keeps doing it for every new one.
- ToolAI PatternsMost teams building agent UI solve the same ten problems from scratch. This is the shared vocabulary — production patterns for agentic UI, each paired with a machine-readable UX contract.