About Me
I’m a data scientist and econometrician working at the intersection of causal inference, experimentation, and machine learning. I build and apply statistical and ML methods to solve practical problems, from production data products to experimentation and decision-making. I’m currently a Senior Data Scientist at Glovo/DeliveryHero.
I have a PhD in Econometrics/Statistics from Universitat Pompeu Fabra in Barcelona. Both in applied and research work, I’m interested in understanding the full consequences of data-driven decisions. My work focuses particularly on going beyond average effects to understand how interventions affect different people and outcomes, and how that information can lead to better decisions.
This website is my personal data hub, where I share my research on causal inference, teaching materials, open-source projects, and writing on data science and statistical methods. Feel free to explore my CV or connect with me via the links on the left.
Research
I develop and apply statistical methods for understanding how interventions affect different people and outcomes, with a particular interest in heterogeneous and distributional effects.
In economics and business, decisions are often made from averages: average treatment effects, average outcomes, average responses. But interventions can affect users very differently, and the parts of the distribution that averages hide can matter for both impact and risk.
My research aims to bring these richer perspectives into practical causal inference, particularly in challenging observational settings where heterogeneity, endogeneity, and other forms of unobserved structure complicate estimation and inference.
See the research page for my work.
Teaching
Over the years, I have taught data science at both undergraduate and graduate levels. My teaching approach emphasizes modern, hands-on learning and good theoretical understanding using open-source tools, reproducible workflows, and accessible explanations. Full course materials, including lecture notes, code, and exercises, are available openly on my GitHub and the teaching page.
Shortcuts to some recent classes with full materials:

Fundamentals of Monte Carlo Simulations in Data Science
Description: Learning the practice of Monte Carlo simulations for evaluating causal, ML, and inference methods.
Access Materials GitHub

Course: Advanced Econometrics (Econometrics II)
Description: Going beyond basics: more theory; causal inference and machine learning basics; more contexts. Empirical examples in Python.
Access Materials GitHub

Course: Econometrics with Unobserved Heterogeneity
Description: A course on methods for estimating parameters of interest in settings with unobserved heterogeneity. Topics include linear models with heterogeneous coefficients, nonparametric models with unobserved heterogeneity, and quantile and distribution regression.
Access Materials GitHub
Blog
I occasionally blog about topics in data science, econometrics, and programming. Posts range from technical walkthroughs to small curiosities I encounter in my work.
Check out the most recent posts:

Introducing unit-averaging
unit-averaging: A package for unit averaging in Python. Efficient ensemble estimation of unit-specific parameters Continue reading Introducing unit-averaging

How to Stand Out in a Master's in Economics Application?
Having credible and verifiable projects is often enough for an outstanding application Continue reading How to Stand Out in a Master's in Economics Application?

The Hidden Delta Method in statsmodels (A Worked Example)
Using the delta method in Python with statsmodels for nonlinear inference: a practical example of confidence intervals and tests Continue reading The Hidden Delta Method in statsmodels (A Worked Example)

Why I Switched from Beamer to Quarto Reveal.js for My Presentations
Why I switched from Beamer to Quarto Reveal.js for reproducible, maintainable, and portable slides in teaching, research, and data science. Continue reading Why I Switched from Beamer to Quarto Reveal.js for My Presentations