I create AI systems that simplify complex workflows.

I don't design for a user — I design for a system: multiple roles, multiple processes, multiple automations. The first thing I map is every actor in the workflow, end to end, before deciding which ones the product is actually for.

That method came out of urban planning, where I was paid to do it before I ever did it for software. In 2015 I mapped a multi-organization cycling network across six public and private entities — a water utility, a power company, traffic police, two federal ministries and civil-society groups — none of them with authority over the next, and I defined the node typology, the phased rollout and the cost model. The same year, as a GIS analyst and cartographer at SEDESOL, Mexico's federal social development ministry, I ran cartography and spatial analysis for federal social programs, digitizing territorial layers and cross-referencing official datasets to support program targeting. Six organizations with no authority over each other is the same problem shape as a multi-role enterprise platform, eleven years earlier. My Master's in Strategic Design and Innovation connected that lens to research, strategy, and implementation, and it still defines how I work today: I look at products as living systems, not isolated screens.

Today, my work sits at the intersection of enterprise workflows, CRM/ERP and cloud-based platforms, design systems shipped as code (tokens, components and a Storybook, not just a Figma library), and AI-assisted decision-making — the same lens that applies whether a job description calls the role Senior UX/UI Engineer, Product Designer, or AI Workflow Engineer. I work in English at full professional proficiency (C1), and Spanish is my native language. I care about AI when it improves judgment, coordination, and execution inside real workflows — not when it becomes decoration. I'm especially interested in products where humans and AI can work together without losing clarity, governance, or control.

I evaluate AI automation across three maturity levels — deterministic workflow automation (L1), agentic integration (L2), and adaptive agent loops (L3) — to decide when automation should stop and human judgment should begin. At Agentic Dream, I applied that framework to Strata, shipping L2 in production where the decision surface was narrow enough to trust, and designing L3 governance patterns — including audit trails and permission scopes — held back until safe to remove human review.

Discovery is the step before the framework. On Repfabric I researched twenty roles and built for one; on Strata, fourteen identified users became the three that drove adoption. I tell you what to build, and I can build it.

I can trace every design decision back to the observation it came from.

That same framework has to hold up at the interface level, not just the visual one. The states an AI-assisted workflow actually needs — is this action machine-initiated, is it generating or did it fail, where did this result come from — live at sebastianbalderas.com/storybook, built on the same design system as the rest of the site.

The same discipline shows up in how I scope. On Strata, research surfaced 14 distinct users and 25+ pain points; I scoped the model down to the 3 driving adoption and designed for their control points instead of trying to solve everything at once. On Repfabric, my research into who actually used the platform versus who bought it brought a 20-role CRM/ERP down to the one persona who drove adoption — a call reached in conversation with the client, weighed against the cost of building for all twenty. Two projects, the same move. That makes it a method rather than a one-off, and it's usually the decision that makes everything after it shippable.

Sole designer across both client engagements. On Avanto/Strata, I presented new work daily to the CEO, project managers and engineers to set priorities and review decisions. On Repfabric, that cadence was near-daily working sessions with founder John Mitchell, who reviewed and approved every design decision.

This site is the proof. I researched, designed, built, deployed and tested sebastianbalderas.com end to end on my own — Next.js, design tokens mapped straight to code, and the conversational layer you're using right now. No engineer, no team. That's my own site, not a platform with a team behind it. The claim isn't that I replace engineering — it's that I can build what I design, well enough to know what I'm asking for.

It's also chat AI first, and that's a product decision rather than a widget: there is no contact form on this site. The assistant is the way in — it answers from my actual work, and it's built to say where I don't fit instead of agreeing with every brief.

If your product needs clearer decisions, safer automation, and workflows that hold up under pressure — you're in the right place.