About

Most AI projects fail not because of the technology, but because nobody read the organization before installing it. I read the human fabric and build the system with the same hand.

I started by studying how people and organizations work, and ended up building the systems, the data and the AI that serve them. Not two careers: one and the same axis.

My foundation is the sociology of organizations and economic analysis in the public sector. From there I moved into data (analysis, modeling, BI), then into data engineering and backend, and today I'm a senior platform engineer (10+ years): I own the infrastructure as code, the secure backends and the pipelines in regulated fintech environments where auditability is a first-class requirement.

I'm not "pivoting" to AI: I already run agentic AI in production. I've built my own MCP servers and auditable controls, and moved up a level toward the infrastructure, control and adoption of AI. I combine platform + data + governance in a single profile. The sociotechnical foundation makes compliance, process modeling and communication with non-technical stakeholders part of the core. That is the ground where Responsible AI and AI Adoption operate.

I work on a single thesis: every organization is a sociotechnical system, so I read the human fabric before touching the technology. The full framework —the bridge thesis, the four-link method and the foundations behind it— is in Approach, and developed in depth in the digital book The Sociotechnical Bridge.

From discovery to proposal, and from there to building

As a consultant I own the full cycle, with no handoffs that degrade meaning. It starts before any code: I sit with the client to understand where it really hurts, reverse-engineer their reporting, design a credible architecture, size it into deliverable packages, and translate the proposal into terms a non-technical decision-maker can act on. Sizing it right up front is what separates a data project that pays off from one that evaporates: most fail from bad scoping, not bad engineering. And because whoever runs discovery is the one who builds, the hypothesis about the organization is kept from day one through to delivery.

Mendoza FuturIA

I founded it and I lead it. Mendoza FuturIA is an independent regional observatory that measures organizational maturity for adopting AI. It didn't come from a client brief: I had the idea and built it to put my own approach to the test —the sociotechnical reading, taken to a concrete instrument—. Its core is a proprietary model, IMIA (7 dimensions × 6 levels), where governance acts as a gate: no organization reaches integration maturity without verifiable AI rules. The snapshot matters, but in SMEs adoption is usually episodic: IMIA is repeated to sustain sociotechnical hygiene, not as proof of arrival. It's one of the projects I show here, not my only face.

Availability

• Languages: Spanish (native), English (C2), Portuguese (bilingual).
• Time zone: UTC-3 — full overlap with the US and European mornings.
• Mode: full-time remote or contractor. Billing via Deel/Ontop.