p[1st surname][2nd surname]@g.harvard.edu
I’m a PhD Candidate at Harvard advised by Elisabeth Paulson and Milind Tambe, and I collaborate closely with David Alvarez-Melis. I study algorithmic decision-making in high-stakes social contexts, especially within public-sector operations. My research is both data-centric—examining how data is collected, labeled, and used in dynamic environments—and decision-oriented, drawing on decision-focused learning to train models that optimize the quality of the decisions they inform rather than only predictive accuracy. From an applications standpoint, I’m currently working with Geomatch through the Stanford Immigration Policy Lab on ML-driven refugee resettlement strategies, and I previously collaborated with ARMANN to design multi-agent–based systems for allocating scarce resources in maternal and child care.
About me. I was born and raised in Bogotá, Colombia, where I obtained a BSc in Mathematics and a BSc and MSc in Industrial Engineering (Operations Research) from Universidad de Los Andes. I am grateful for how the community I grew up in has shaped me, and I strive to contribute meaningfully to my Latin American community through my research, mentoring, and service. In this spirit, I co-founded and continue to help lead the Centro de Analítica para Políticas Públicas (CAPP), an initiative dedicated to facilitating research in data science and AI to address the most pressing challenges faced by the public sector in Colombia and Latin America. I am also involved with the EAAMO community and used to co-organize the ML4D Workshop at NeurIPS.
News
November 2025
October 2025
Giving an invited talk in the Data-Driven Sustainable Operations session at INFORMS 2025 in Atlanta. (Sunday, Oct 26 at 2:45pm)
June 2025
Our paper What is the Right Notion of Distance between Predict-then-Optimize Tasks? was accepted at UAI 2025
October 2024
Attending EAAMO Conference in San Luis Potosí, Mexico. Excited to serve as a poster co-chair for the conference! (Oct 29-31)
October 2024
September 2024
In collaboration with EAAMO and CAPP, we organized the colloquium “Governing Colombia’s Biocultural Diversity: A Holistic Approach” as part of EAAMO’s first colloquium series focused in the Latin America and the Caribbean (LAC) region. [talk recording] [discussion panel recording]
September 2024
September 2024
Started as a Teaching Fellow (Teaching Assistant) for Harvard’s CS2822R Topics in Machine Learning: Computational Properties in Interpretable Machine Learning, taught by Finale Doshi-Velez
July 2024
Attending ICML in Vienna. I’ll be presenting our work on Dataset Distance for Predict-then-Optimize Tasks at DMLR and Humans, Algorithmic Decision-Making and Society Workshops and serving as a volunteer for WiML. (July 21-27)
October 2023
Serving as a poster co-chair for EAAMO Conference, hosted at Boston University. (Oct 30-Nov 1)
October 2023
Attending INFORMS in Phoenix, AZ. I was invited to give a talk at the session on Fairness in Decision-Making. (Oct 15-18)
September 2023
Started as an instructor for Harvard’s CS290, Seminar on Effective Research Practices & Academic Culture
June 2023
Visited our collaborators at ARMMAN in India to learn more about their mHealth program focused on improving maternal and child health.
May 2023
Attending AAMAS in London. I’ll give an oral presentation on our paper Fairness for Workers Who Pull the Arms: An Index Based Policy for Allocation of Restless Bandit Tasks (May 30)
April 2023
I successfully passed my qualifying exam and am now officially a PhD candidate!
February 2023
Attending AAAI in Washington D.C. I’ll be giving an oral presentation on our paper Flexible Budgets in Restless Bandits: A Proximal Primal-Dual Algorithm for Efficient Budget Allocation and co-organizing the Women’s Mentoring Lunch (Feb 7-14)
Paula Rodriguez-Diaz 
