Mateo Dulce Rubio

¡Hola! I am an Assistant Professor & Faculty Fellow at the NYU Center for Data Science. My research develops flexible and robust statistical tools that turn machine learning predictions of unknown quality into decisions with provable guarantees, together with the conditions a predictive model must satisfy for these guarantees to hold. This agenda combines methods from nonparametric statistics, causal inference, optimization, and machine learning, and is developed for humanitarian and policy applications in close partnership with organizations such as UNOPS and The HALO Trust. I'm broadly interested in:

• Algorithmic decision-support systems for landmine clearance, motivated by operational and data challenges in humanitarian mine action. These systems have been used in real-world operations in Colombia and Afghanistan, and have been recognized by INFORMS, GICHD, ITU, and Kluz PeaceTech.
• Doubly robust estimators from ML predictors for causal inference and population size estimation with heterogeneous capture probabilities and recapture dependencies, with applications in conservation, public health, and human rights.
• Design and deployment of AI systems (e.g., tabular or language foundation models) for resource allocation under uncertainty, how to incorporate them into broader decision-making workflows with honest guarantees, and how to audit their behavior once deployed.

I hold a PhD in Statistics and Public Policy from Carnegie Mellon University, where I Was advised by Edward H. Kennedy and received the William W. Cooper Doctoral Dissertation Award for my thesis “Robust Nonparametric Methods for Peacebuilding.”

I am on the 2026–2027 academic job market.

* Dulce is my first (paternal) last name and Rubio is my second (maternal) last name.