I’m Carlos Daniel Jiménez. I build AI systems that work in production—recommendation engines, intelligent agents, and the MLOps infrastructure that keeps them running.
My focus: Agentic AI and MLOps. Edge Computing is my research playground.
The path here
Throughout my journey in AI, I’ve had the privilege of working on problems that actually matter—from detecting fraud patterns at Mercado Libre to evaluating social development projects across Latin America at the Inter-American Development Bank.
Currently Principal MLOps Engineer at Globant, where I design ML pipelines and AI solutions for clients in the UAE and United States. My day-to-day: recommendation systems, LLM-powered applications, computer vision, chatbots, and making sure models don’t just work in notebooks.
Before Globant, I was AI Engineer at Yale University, collaborating with NVIDIA on edge computing research—optimizing memory usage, energy consumption, and model weights for constrained devices. That project solidified something I already suspected: the future of ML isn’t just in the cloud.
At Mercado Libre, I led a team building pricing models for Argentina, Brazil, and Mexico. When your Bayesian model affects millions of dynamic auctions, you learn to care deeply about uncertainty quantification.
Earlier: ML Engineer at the Inter-American Development Bank (AI for policy evaluation), Data Scientist at the United Nations in Chile (criminal detection models), and five years at Bogotá’s Education Department building the first AI system that automatically wrote institutional reports.
What I care about
Agentic AI — Systems that don’t just predict, but act. I’m interested in how we build AI that makes autonomous decisions, and the infrastructure required to deploy and monitor it reliably.
Edge Computing — This is where I experiment. I build computer vision models on Raspberry Pi and Jetson Nano, optimizing for memory and energy. It’s ML where cloud isn’t an option.
Statistical rigor — I came from economics and philosophy, not computer science. I still think most ML practitioners underestimate uncertainty and overestimate their models.
Teaching
I’ve taught Machine Learning, Applied Statistics, and Bayesian methods in Colombia, Mexico, Peru, Chile, and Spain. Teaching is how I stress-test my own understanding—if I can’t explain it clearly, I probably don’t understand it well enough.
A fun fact
I’m pursuing a Master’s in Economic History, researching the influence of religion on economic development. It sounds unrelated to AI, but it’s not: both fields are about understanding systems, incentives, and why humans behave the way they do.
I collect vinyl records. I also build and repair guitars—Jazzmaster-style, because their floating tremolo and offset body are the most interesting engineering problem in luthería.
Let’s connect
📬 I write about MLOps, Agentic AI, and Edge Computing. Subscribe to the newsletter →
📍 Based in Colombia, working globally.