PeopleofSuperintelligence

Researcher

Elena Buchatskaya

Elena Buchatskaya is an AI researcher affiliated with Google who has co-authored papers on self-supervised learning and large language models since 2020. She was an author of BYOL in 2020, of the Gopher language model analysis in 2021, and of the Chinchilla paper in 2022, which found that model size and training tokens should be scaled equally and won an Outstanding Paper award at NeurIPS 2022. Her name also appears on the Gemini report in 2023, the Gemini 1.5 and Gemma reports in 2024, and the Gemma 3 and Gemini 2.X reports in 2025. In October 2026, Nature Medicine published the MedGemma paper, a collection of medical vision-language models based on Gemma 3, with her among its authors.

Now at
Google

Timeline

  1. 2020

    Self-supervised learning with BYOL

    In June 2020, she is among the authors of Bootstrap Your Own Latent (BYOL), a self-supervised method for learning image representations. BYOL achieves a new state of the art without the negative pairs that other state-of-the-art methods rely on.

  2. 2021

    The Gopher language model

    In December 2021, she is among the authors of "Scaling Language Models: Methods, Analysis & Insights from Training Gopher." The paper analyzes Transformer language models up to Gopher, a 280 billion parameter model, on 152 diverse tasks.

  3. 2022

    Routed scaling laws and Chinchilla

    In February 2022, she is among the authors of "Unified Scaling Laws for Routed Language Models," which studies Routing Networks that use only a subset of their parameters for each input. In March 2022, she is among the authors of the Chinchilla paper, which finds that for compute-optimal training, model size and training tokens should be scaled equally: every doubling of model size calls for a doubling of training tokens.

  4. 2022

    A NeurIPS Outstanding Paper award

    In 2022, the Chinchilla paper, titled "An empirical analysis of compute-optimal large language model training" at NeurIPS, wins an Outstanding Paper award, and she is one of its authors.

  5. 2023

    The Gemini technical report

    In December 2023, she is among the authors of the Gemini report, which introduces the multimodal Gemini family in Ultra, Pro and Nano sizes.

  6. 2024

    Gemini 1.5 and Gemma

    In March 2024, she is among the authors of the Gemini 1.5 report, which describes models capable of recalling and reasoning over fine-grained information from millions of tokens of context. She is also among the authors of Gemma, a family of lightweight open models built from the research and technology used to create Gemini.

  7. 2025

    Gemma 3 and Gemini 2.5

    In March 2025, she is among the authors of the Gemma 3 report, which adds vision understanding and a context of at least 128K tokens to the open Gemma models. In July 2025, she is among the authors of the Gemini 2.X report, which introduces Gemini 2.5 Pro and Gemini 2.5 Flash.

  8. As of October 2026

    MedGemma in Nature Medicine

    In July 2025, she is among the authors of the MedGemma Technical Report, which introduces medical vision-language foundation models based on Gemma 3. In October 2026, Nature Medicine publishes the article "An open vision-language model for diverse medical applications," and she is among its authors, with a Google affiliation.

  1. 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.
  2. 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.
  3. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes Google. 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.
  4. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Google. 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.

Sources and credits

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

  1. Bootstrap your own latent: A new approach to self-supervised Learning arxiv.org. Checked 21 October 2026.
  2. Scaling Language Models: Methods, Analysis & Insights from Training Gopher arxiv.org. Checked 21 October 2026.
  3. Unified Scaling Laws for Routed Language Models arxiv.org. Checked 21 October 2026.
  4. Training Compute-Optimal Large Language Models arxiv.org. Checked 21 October 2026.
  5. NeurIPS 2022 Awards neurips.cc. Checked 21 October 2026.
  6. Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 21 October 2026.
  7. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context arxiv.org. Checked 21 October 2026.
  8. Gemma: Open Models Based on Gemini Research and Technology arxiv.org. Checked 21 October 2026.
  9. Gemma 3 Technical Report arxiv.org. Checked 21 October 2026.
  10. Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities arxiv.org. Checked 21 October 2026.
  11. MedGemma Technical Report arxiv.org. Checked 21 October 2026.
  12. An open vision-language model for diverse medical applications | Nature Medicine nature.com. Checked 21 October 2026.