PeopleofSuperintelligence

Lab leader

Jonathon Shlens

Also known as Jon Shlens

Jonathon Shlens is an AI researcher whose work links biological vision with computer vision and machine-learning systems. He earned a PhD in computational neuroscience at UC San Diego in 2007 and joined Google Research in 2010 after research roles at the Howard Hughes Medical Institute, Pixar Animation Studios and UC Berkeley. His work includes the 2015 adversarial-examples paper, the TensorFlow system paper, multimodal robustness research, MOFI and a visual memory model. As of October 2026, he is a senior director of research and distinguished scientist at Google DeepMind, with interests in vision, language and learning.

Badges

Timeline

  1. 2007

    A computational neuroscience PhD

    He earns a PhD in computational neuroscience at UC San Diego in 2007. His research applies machine learning to visual processing in biological systems.

  2. 2010

    Google Research and vision systems

    Before Google, he works as a research fellow at the Howard Hughes Medical Institute, a research engineer at Pixar Animation Studios and a Miller Fellow at UC Berkeley. He joins Google Research in 2010 and focuses on scalable vision systems.

  3. 2015

    Explaining adversarial examples

    With Ian Goodfellow and Christian Szegedy, he coauthors “Explaining and Harnessing Adversarial Examples” in 2015. The paper traces neural networks’ vulnerability to adversarial perturbations to their linear nature and offers a fast method to generate examples.

  4. 2015–2016

    TensorFlow’s system paper

    In November 2015, the TensorFlow API and reference implementation become an open-source package under Apache 2.0. In 2016, he coauthors the system paper, which describes a machine-learning interface that runs from mobile devices to large distributed systems.

  5. 2023

    Multimodal robustness

    In 2023, he coauthors a study of multimodal learning under train-test modality mismatch. The study proposes two interventions that improve robustness by 1.5×–4× across AudioSet, Kinetics-400 and ImageNet-Captions.

  6. 2024

    MOFI and the Google move

    In 2024, he coauthors MOFI, a vision foundation model that learns image representations from noisy entity-annotated images. By May 2024, he works at Google after contributing to MOFI while at Apple.

  7. 2025

    A visual memory system

    In 2025, he coauthors a visual memory system that combines pretrained image embeddings with nearest-neighbor retrieval. The system can add individual examples and entire classes and remove data through unlearning and memory pruning.

  8. As of October 2026

    Google DeepMind research director

    As of October 2026, he is a senior director of research and distinguished scientist at Google DeepMind. His interests include vision, language and learning.

Career

  1. NowGoogle DeepMindSenior director of research and distinguished scientist
  2. Until 2024Apple
  3. From 2010Google Research
  4. Howard Hughes Medical InstituteResearch Fellow
  5. Pixar Animation StudiosResearch Engineer
  6. University of California, BerkeleyMiller Fellow

Education

  1. 2007University of California, San DiegoPhD

From the biography.

  1. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Jonathon Shlens. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.
  2. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Jonathon Shlens. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  3. Google had the most people shaping AI every year from 2014 to 2025Includes Google. Our estimates of which organizations employed the most people shaping AI each year from 1955 to 2025, tracing the shift from university labs to companies.
  4. Google has the highest combined h-index among AI organizationsIncludes Google. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.

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 Jonathon Shlens yet. Send us a free-licensed photo.

  1. Learning representations of the visual world | Fields Institute for Research in Mathematical Sciences fields.utoronto.ca. Checked 21 October 2026.
  2. Explaining and Harnessing Adversarial Examples research.google. Checked 21 October 2026.
  3. Citing TensorFlow | TensorFlow tensorflow.google.cn. Checked 21 October 2026.
  4. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems arxiv.org. Checked 21 October 2026.
  5. Robustness in Multimodal Learning under Train-Test Modality Mismatch machinelearning.apple.com. Checked 21 October 2026.
  6. MOFI: Learning Image Representations from Noisy Entity Annotated Images arxiv.org. Checked 21 October 2026.
  7. International Conference on Learning Representations (ICLR) 2024 machinelearning.apple.com. Checked 21 October 2026.
  8. Towards flexible perception with visual memory proceedings.mlr.press. Checked 21 October 2026.
  9. Jonathon Shlens shlens.github.io. Checked 21 October 2026.