AI Enterprise Architect & Governance — Enterprise Architecture
I'm from Bogotá. I've spent years in Ecuador —Quito, now Cuenca—; before that I lived in Australia for a while.
I work at the boundary where technology, people, and AI meet: I step into organizations that have outgrown their own systems, understand their internal logic, and design the architecture —and the AI governance— that lets them keep growing without breaking.
Today I'm building the enterprise-AI capability of a consumer-goods multinational: solution architecture, integrating GenAI and agents into the business, and end-to-end AI-First adoption.
One focus, in depth: Enterprise AI Architecture and Governance. I don't train models or run scattered experiments — I stand up the function, earn executive sponsorship, govern how AI and agents are used across the company, and turn scattered use cases into a real, adopted capability.
- Architecture of an end-to-end commercial transformation for a consumer-goods multinational — TO-BE model, GAP analysis, systems integration
- A generative-AI pipeline for meeting transcription and minutes (Whisper + diarization + cleanup)
- Reunia — an AI-first meeting-intelligence platform (in progress)
- LLM-agent workflows that produce documentation and deliverables consistently
- ERP deployed on the cloud with Terraform and CI/CD; executive dashboards (Apache Superset) for budget vs. actual; automation that removed around 70% of manual work
- Architectural modeling and documentation (C4/Structurizr, BPMN, ArchiMate, ADRs) where none existed before
These aren't lab projects. They're systems someone uses today to make decisions that cost real money.
When I step into a complex organization, the first thing isn't code — it's understanding. I map out processes by talking to people, model them formally (C4, BPMN, ArchiMate), and record the decisions that matter as ADRs, so no one has to guess later why something was done. I treat governance the same way: clear rules for how AI and data are used, not locks that slow the business down.
I treat documentation as code: versioned, reviewable, alive. And when a repetitive task can be automated —from transcribing meetings with AI to generating deliverables— I automate it, so time gets spent thinking, not copying.
AI & GenAI OpenAI · Anthropic (Claude) · Vertex AI / Gemini · RAG · Agents · Whisper
Governance & Risk NIST AI RMF · ISO/IEC 42001 (adopting) · DLP · AI usage policies
Enterprise Architecture C4 / Structurizr · BPMN 2.0 · ArchiMate · ADRs · TOGAF (in progress)
ERP & Data Odoo (v14–v20) · PostgreSQL 16 · OCA Modules · Python
Cloud & DevOps GCP · AWS · Docker · Nginx · Terraform · CI/CD
Observability Prometheus · Grafana · Loki · cAdvisor
Business Intelligence Apache Superset · Metabase · SQL Views
Docs & Automation Obsidian · LaTeX · React / TypeScript
PENTA Digital Lab · AI & ERP Solutions Architect — May 2026 – present · Cuenca, Ecuador Enterprise architecture, GenAI and agent integration, and AI-First adoption for a consumer-goods multinational.
Oil & Gas Engineering · Solutions Architect & Process Analyst — 2025 – 2026 Enterprise architecture, ERP, BI, and digital transformation for distributed engineering operations across LATAM.
Independent consulting · Solutions Architect — 2020 – 2024 Odoo implementations, custom modules, and digital-transformation proposals for regional companies.
NUTRABIOTICS S.A.S · Full Stack Developer — 2022 – 2023 Odoo development, information security, and AWS infrastructure.
Earlier — education, engineering, financial technology.
Where I learned that if you can't explain something clearly, you probably don't understand it well enough yet.
Systems Engineering — UNAD Colombia (in progress) 47+ certifications in cloud, security, data, and architecture Currently pursuing TOGAF and AI-governance frameworks (NIST AI RMF, ISO/IEC 42001)
I want to lead AI architecture and governance in international organizations, remotely. I'm not driven by the title — I'm driven by building AI capabilities that last and that matter.