PeopleofSuperintelligence

Researcher

Juan Perdomo

Also known as Juan Carlos Perdomo, Juan C. Perdomo, Juan Carlos Perdomo Silva

Juan Perdomo is a computer scientist who studies what happens when machine learning predictions guide decisions about people. A sailor from San Juan, Puerto Rico, he won the island's first gold medal at the 2013 Youth World Championships before turning to research. In 2020 he coauthored the ICML paper Performative Prediction, and in 2023 he completed a PhD in EECS at UC Berkeley, advised by Peter Bartlett and Moritz Hardt. He held postdoctoral posts at Harvard's Center for Research on Computation and Society and at MIT, won an ICML 2025 Outstanding Paper Award for The Value of Prediction in Identifying the Worst-Off, and was named a 2026 AI2050 Early Career Fellow. He is an assistant professor of computer science and data science at New York University.

Timeline

  1. 2013–2015

    Sailing for Puerto Rico

    A sailor from San Juan, he wins the island's first gold medal at the 2013 Youth World Championships and represents Puerto Rico at the 2015 Pan American Games. He sails for the national team before turning to research.

  2. 2020

    Performative Prediction at ICML

    With Tijana Zrnic, Celestine Mendler-Dünner and Moritz Hardt, he coauthors Performative Prediction at the 2020 International Conference on Machine Learning. The paper develops a risk-minimization framework for predictions that influence the outcomes they aim to predict, defines performative stability, and gives conditions for retraining to converge to a performatively stable point.

  3. September 2022

    Wisconsin's dropout warning system

    As a UC Berkeley EECS graduate student, he presents an evaluation of the early warning system designed by the Wisconsin Department of Public Instruction at the Berkeley Evaluation & Assessment Research Center in September 2022. Drawing on nearly a decade of data, the study finds the system's risk assessments highly accurate and sees no evidence that its predictions have meaningfully changed graduation outcomes.

  4. May 2023

    A Berkeley EECS PhD

    He completes his PhD in EECS at UC Berkeley in May 2023 with the thesis Performative Prediction: Theory and Practice, advised by Peter Bartlett and Moritz Hardt. Before Berkeley, he earns a BA in computer science and mathematics at Harvard and works briefly as a data scientist at Uber.

  5. April 2024

    Postdoc at Harvard's CRCS

    As a postdoctoral fellow at Harvard's Center for Research on Computation and Society, hosted by Cynthia Dwork, he gives a CITP seminar at Princeton in April 2024 titled The Relative Value of Prediction. The talk compares the welfare benefits of improving prediction with those of expanding access to social goods and presents a case study of the risk prediction system used across Wisconsin public schools.

  6. 2025

    An ICML Outstanding Paper Award

    With Unai Fischer Abaigar and Christoph Kern, he wins an Outstanding Paper Award at ICML 2025 for The Value of Prediction in Identifying the Worst-Off, one of six outstanding papers at the conference. The paper compares the welfare effects of better prediction with other policy levers such as expanding bureaucratic capacity, using mathematical models and a case study of long-term unemployment among German residents.

  7. 2026

    An AI2050 Early Career Fellowship

    In 2026, Schmidt Sciences names him one of 22 early-career fellows in its fifth cohort of AI2050 fellows. His project develops the theoretical tools and practical frameworks that keep AI-assisted allocation systems in public institutions efficient, fair and trustworthy, and carries them into deployments with partner government agencies.

  8. Since September 2026

    Assistant professor at NYU

    After postdoctoral fellowships at Harvard and MIT, he starts as an assistant professor of computer science and data science at New York University in September 2026. His research studies the feedback loops between people and algorithms and develops methods with provable guarantees in those social settings.

Career

  1. Since 2026New York UniversityAssistant professor
  2. UberData scientist
  3. Massachusetts Institute of TechnologyPostdoctoral fellow

Education

  1. 2023University of California, BerkeleyPhD
  2. Harvard UniversityBA

From the biography.

  1. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Juan Perdomo. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  2. The median h-index is 48 at universities and 14 at frontier AI labsIncludes New York University. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  3. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes New York University. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  4. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes New York University. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Juan Perdomo yet. Send us a free-licensed photo.

  1. Juan Carlos Perdomo – Hire Aspirations Institute hireaspirations.seas.harvard.edu. Checked 21 October 2026.
  2. Performative Prediction proceedings.mlr.press. Checked 21 October 2026.
  3. Juan Perdomo: Can Prediction Systems Help Reduce High School Dropout? A Retrospective on Wisconsin Public Schools | Berkeley Evaluation & Assessment Research Center bearcenter.berkeley.edu. Checked 21 October 2026.
  4. Performative Prediction: Theory and Practice | EECS at UC Berkeley www2.eecs.berkeley.edu. Checked 21 October 2026.
  5. Juan Perdomo simons.berkeley.edu. Checked 21 October 2026.
  6. CITP Seminar: The Relative Value of Prediction | Princeton School of Public and International Affairs spia.princeton.edu. Checked 21 October 2026.
  7. ICML 2025 Awards icml.cc. Checked 21 October 2026.
  8. Congratulations to the #ICML2025 award winners! - AIhub aihub.org. Checked 21 October 2026.
  9. Schmidt Sciences Names 28 AI2050 Fellows Tackling the Defining AI Challenges of Our Time - Schmidt Sciences schmidtsciences.org. Checked 21 October 2026.
  10. Juan Carlos Perdomo - AI2050 - Schmidt Sciences ai2050.schmidtsciences.org. Checked 21 October 2026.
  11. FDS Colloquium: Juan Carlos Perdomo Silva (NYU), "The Relative Value of Prediction" - Yale FDS fds.yale.edu. Checked 21 October 2026.
  12. Juan Carlos Perdomo jcperdomo.org. Checked 21 October 2026.
  13. NYU Computer Science Department cs.nyu.edu. Checked 21 October 2026.