About

My first contact with programming happened inside Minecraft. I already played it as a kid, but as a teenager I drifted toward the more technical side of the game - I started experimenting with commands, building custom items and mobs and sketching out a Capture-the-Flag-style map that never quite got finished. That was the moment the word programming stopped being abstract for me. I went looking for “how to learn to code” and landed on the usual web stack - HTML, CSS and JavaScript - then moved into Python, a language I had been hearing about for years, and followed it into machine learning once I saw what it was actually used for.

I'm Hernan Rochon, a ML Engineer and Data Scientist based in Uruguay. I build end-to-end machine learning systems that solve concrete business problems - customer risk, churn, anomaly detection, demand anticipation - turning them into reliable services that decision-makers can actually act on. The work spans the full ML lifecycle, from data pipelines and feature engineering to model training, evaluation and deployment, with a strong focus on production readiness, interpretability and measurable impact.

I care about systems that are reproducible, testable and honest about their limitations - models that ship, keep working and tell you when they shouldn't be trusted.

I'm currently looking for my first professional role as a ML Engineer or Data Scientist. I'm open to remote opportunities.