I'm an Adjunct Professor of Mathematics at Universitat Pompeu Fabra and of Statistics at BarcelonaTech (UPC). Prior to this, I earned my Ph.D. in Statistics at the University of Wisconsin - Madison as a "la Caixa" Fellow, advised by Professor Wei-Yin Loh. My focus is on trustworthy AI, and I'm currently developing TRUST (Transparent, Robust and Ultra-Sparse Trees) - the most interpretable model tree algorithm ever created. Our results show that already earlier versions of TRUST often matched or exceeded the predictive accuracy of popular black-box machine learning models like Random Forests, while remaining fully explainable.

My goal is simple: to provide innovative and safe AI tools that allow users in high-stakes domains to stop choosing between accuracy and interpretability - and, in doing so, make a positive impact on society. In a world that is being pushed towards ever-increasing complexity and opacity, I am instead redefining the boundaries of what white-box models for tabular data can accomplish.

Before joining UW-Madison, I worked as a financial risk analyst at the European Central Bank. Earlier, I was a Master's student at BarcelonaTech (UPC), an exchange student-athlete at Carnegie Mellon University and a double-degree undergraduate student at Universitat Pompeu Fabra (UPF).

Here on GitHub, I version-control the latest developments related to my TRUST algorithm. My Python TRUST package trust-free is hosted on PyPI and can be downloaded for free and installed via pip install trust-free.

trust-free is a Python package for fitting interpretable regression models using Transparent, Robust, and Ultra-Sparse Trees (TRUST) — a new generation of Linear Model Trees (LMTs) with high accuracy and intuitive explanations. It is based on my peer-reviewed paper, recently published as PRICAI 2025 proceedings on Lecture Notes in Artificial Intelligence (Springer Nature).

The package currently supports standard regression and experimental time-series regression tasks. Future releases will also tackle other tasks such as classification.