Hi, I’m Mateo.

I build ML systems for complex environments.

I’m a Senior Machine Learning Engineer at Vu.City and an Honorary Research Fellow at UCL.

I specialize in spatial and graph-structured data, moving between academic research and production engineering. My work focuses on turning high-dimensional, messy data into models and systems that can survive contact with the real world.

I have a research background in Network Science and Urban Analytics (PhD, UCL / Alan Turing Institute), where I studied how network structure and information shape human behaviour in cities. Since then, I’ve spent much of my time bridging the gap between research and production: designing, building, and deploying data and ML systems inside practice-driven environments.

I have spent my career applying that rigor to industry. Previously, as an Associate Partner at Foster and Partners, I led the deployment of data-driven tools that influenced urban design at scale. Before that, I developed transport algorithms at SignalBox.

My work is defined by synthesis: taking ideas from papers, prototypes, or whiteboards and translating them into robust, scalable systems across data pipelines, modelling, and product integration. I’m comfortable operating end-to-end, from experimental work to engineering decisions and stakeholder alignment.

I’m especially interested in problems involving:

• spatial and graph-structured data
• long feedback loops and real-world constraints
• systems where technical choices meaningfully shape decisions