PeopleofSuperintelligence

Researcher

Finbarr Timbers

Finbarr Timbers is an AI researcher whose work spans strategic game agents and open language models. He studied mathematics and economics at the University of Alberta, pursued econometrics at the London School of Economics, and turned to machine learning in 2014. His work includes DeepNash and Student of Games; in 2024, he wrote Artificial Fintelligence and appeared at Upper Bound as a Midjourney speaker. He later coauthored OLMo 3 and Olmo Hybrid, then said his last day at Ai2 was in June 2026.

Links
Website

Timeline

  1. 2014

    From economics to machine learning

    He studies mathematics and economics at the University of Alberta and pursues a master's in econometrics at the London School of Economics. In 2014, he decides to move into machine learning.

  2. 2020

    Approximate exploitability

    In 2020, he coauthors ISMCTS-BR, a scalable deep reinforcement learning algorithm that learns a best response to approximate an agent's worst-case performance.

  3. 2022

    DeepNash in Stratego

    In 2022, he coauthors DeepNash, a Stratego agent that combines game theory with model-free deep reinforcement learning. Against top expert human players on Gravon, DeepNash achieves an 84% win rate and an all-time top-three ranking.

  4. 2023

    Student of Games

    In 2023, he coauthors Student of Games, a general-purpose algorithm combining guided search, self-play learning, and game-theoretic reasoning. It achieves strong results in chess and Go.

  5. 2024

    A Midjourney speaking role

    In 2024, he writes Artificial Fintelligence as an AI researcher with prior experience at DeepMind and Midjourney. At Upper Bound in May 2024, he appears on the speaker list with Midjourney.

  6. 2025

    The Olmo 3 model family

    In 2025, he coauthors OLMo 3, a family of fully open language models at the 7B and 32B scales.

  7. April 2026

    Olmo Hybrid

    In April 2026, he coauthors Olmo Hybrid, which outperforms Olmo 3 on standard pretraining and mid-training evaluations.

  8. June 2026

    Leaving Ai2

    In June 2026, he says it is his last day at Ai2, where he works on OLMo post-training recipes.

  1. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI from 1961 to 2025, with a year-by-year chart.
  2. The United States employs an estimated 53% of people in our data with a known employment countryOur estimates of where AI careers are concentrated now and how countries’ shares of dated positions changed over time.
  3. 46% of living people in our data with known birth and employment countries work outside their birth countryOur estimate of cross-border careers in AI, from country of birth and the country of each current organization.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionOur estimates of where the people shaping AI studied and where their careers led, from education records in cited biographies.

Sources and credits

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

  1. Interviewing Finbarr Timbers on the "We are So Back" Era of Reinforcement Learning interconnects.ai. Checked 19 October 2026.
  2. Approximate exploitability: Learning a best response in large games arxiv.org. Checked 19 October 2026.
  3. Mastering Stratego, the classic game of imperfect information deepmind.google. Checked 19 October 2026.
  4. Student of Games: A unified learning algorithm for both perfect and imperfect information games pmc.ncbi.nlm.nih.gov. Checked 19 October 2026.
  5. Amii’s Upper Bound conference invites all to experience the future of AI kommunikasjon.ntb.no. Checked 19 October 2026.
  6. Olmo 3 arxiv.org. Checked 19 October 2026.
  7. Olmo Hybrid: From Theory to Practice and Back arxiv.org. Checked 19 October 2026.
  8. Frontier post-training recipe review with Finbarr Timbers interconnects.ai. Checked 19 October 2026.