I've spent 15+ years in operations across heavily regulated industries. These days, I build AI systems that actually work in production, not just in demos.
I studied Industrial Engineering at Universidad de los Andes and got my MBA at EAE Business School in Madrid. Over 15 years I've worked in agroindustry, pharmaceutical cannabis, glass manufacturing, and automotive. Different worlds, same core problem: how do you get complex operations to run better?
Three years ago I went all-in on AI. Not the hype, the real thing: computer vision models running on factory floors, WhatsApp agents that farmers actually use, automation that replaced hours of manual SAP data entry. Everything I build has to connect to a P&L somewhere.
In 2025 I started Onza AI to help other companies in Colombia and LATAM do the same. I also run AI at Oleoflores, where we have 4 palm oil extraction plants and a lot of interesting problems to solve.
Every project follows this path. No shortcuts.
What's the actual business problem? I talk to operators, look at the data, sit in on the process. No solutioning yet.
Where does AI actually help vs. where a spreadsheet fix is enough? I map data sources, gaps, and integration points.
Pick the right stack. Design the pipeline. Define how it connects to ERP, BI, or messaging. Plan for what breaks.
Prototype fast, test with real users, iterate. Deploy to production with monitoring. I don't hand off decks, I hand off working systems.
These are real systems running in production, not proofs of concept.
AI-driven system orchestrating fruit and process flows across 4 extraction plants, integrated with SAP/ERP for real-time visibility.
CV models to classify palm fruit quality and detect under-processed bunches, connected to quality dashboards and operational alerts.
AI agents on WhatsApp supporting 500+ farmers with logistics, quality guidance, and agronomic recommendations.
15+ automated workflows connecting SAP, logistics, messaging, and reporting tools to eliminate manual tasks.
Interactive dashboards with AI-generated insights for leadership, tracking supply chain, plant performance, and commercial KPIs.
End-to-end AI solutions for LATAM companies: RAG systems, conversational agents, predictive dashboards, and AI-powered websites.
What I use day to day.
I started Onza AI because I kept seeing the same problem: companies want to use AI but don't know where to begin. We fix that.
Executive and team training on AI adoption, responsible AI, and practical use cases. Delivered for Grupo Mexico and Colombian enterprises.
Workflow automation connecting ERP, logistics, messaging, and reporting tools using n8n and custom integrations.
RAG systems, conversational agents, predictive dashboards, and AI-powered websites. Python, React, LLMs, APIs.
AI adoption roadmaps, use case evaluation, financial impact prioritization, and change management for organizations.
Most AI training is too technical for the people who need it most. I've given 5+ webinars to corporate audiences in Mexico (Grupo Mexico, among others) and Colombia, specifically for non-technical teams. At Oleoflores, I run internal sessions for operations, logistics, and purchasing.
MBA — EAE Business School, Madrid (2015)
Industrial Engineering — Universidad de los Andes, Bogota (2010)
Languages: Spanish (Native) | English (C1) | Portuguese (Conversational)
I'm open to consulting projects, speaking gigs, or just a good conversation about AI in industry.