Profile

I am a Research Scientist at Google working on data privacy. Before joining Google, I completed my Ph.D. in Electrical and Computer Engineering at UT Austin, where I was advised by Prof. Haris Vikalo.

My research focuses broadly on enabling private machine learning systems. Recently I have been particularly interested in empirical privacy tests, to quantify effective risk of differentially private algorithms.

Earlier, I earned my B.S. in Mathematics from Universidad de Los Andes in Bogotá, Colombia, under the supervision of Prof. Mauricio Velasco, and worked as a data science researcher at Quantil.

Selected Publications

A list of publications from 2024 and onwards. For a complete list of my work, please visit my Google Scholar profile.

• Regularized $f$-Divergence Kernel Tests. AISTATS 2026.
  With Antonin Schrab and Arthur Gretton.
  [Paper]

• Sequentially Auditing Differential Privacy. NeurIPS 2025.
  With Tomás González, Mateo Dulce-Rubio, and Aaditya Ramdas.
  [Paper]

• Privately Counting Partially Ordered Data. ICLR 2025.
  With Matthew Joseph and Alexander Yu.
  [Paper]

• Differentially private optimization for non-decomposable objective functions. ICLR 2025.
  With Weiwei Kong and Andres Munoz Medina.
  [Paper]

• DP-Auditorium: A large-scale library for auditing differential privacy. IEEE S&P 2024.
  With William Kong, Andres Munoz Medina, and Umar Syed.
  [Paper]

• Federated learning at scale: Addressing client intermittency and resource constraints. IEEE JSTSP 2024.
  With Haris Vikalo and Gustavo de Veciana.
  [Paper]

Other

Outside of research, I enjoy movement (climbing, yoga, swimming, biking), reading fiction (Annie Dillard’s, Chesterton, Flannery O’Connor stories or Juan Gabriel Vasquez novels, and recently The Brothers Karamazov), and try my best to stay connected with the machine learning community in Colombia and South America.