PeopleofSuperintelligence

Researcher

Sebastian Borgeaud

As of October 2026, Sebastian Borgeaud is a research engineer at Google DeepMind, and his papers study how large language models are trained and scaled. He studied computer science at Cambridge from 2014 to 2018 and joined DeepMind as a research engineer in 2018. In 2021 he co-authored the RETRO retrieval paper and the Gopher training study. The Chinchilla paper, which he co-authored, received a NeurIPS Outstanding Paper award in 2022. He was text co-lead on the Gemini family of multimodal models in December 2023, and in April 2026 a report named him team leader of a Google DeepMind effort on AI coding models.

Works
25
Citations
6,616
h-index
20

Publication numbers from OpenAlex, as of 15 October 2026.

Timeline

  1. 2014–2018

    Computer science at Cambridge

    He studies computer science at Cambridge from 2014 to 2018 and completes a BA/MEng there.

  2. 2018

    Joining DeepMind

    He joins DeepMind as a research engineer in 2018.

  3. 2021

    RETRO and Gopher

    In 2021 he co-authors RETRO, which improves language models by retrieving from trillions of tokens, and the Gopher study on scaling language models.

  4. 2022

    A NeurIPS Outstanding Paper

    The Chinchilla paper, which he co-authors, wins a NeurIPS Outstanding Paper award in 2022.

  5. December 2023

    Text co-lead for Gemini

    He is text co-lead for the Gemini family of multimodal models, whose paper is posted in December 2023.

  6. April 2026

    A strike team on coding models

    In April 2026, Google DeepMind forms a team to improve its AI coding models. The team's leader is Sebastian Borgeaud, a former pre-training lead for Gemini.

  7. As of October 2026

    Research engineer at DeepMind

    He is a research engineer at Google DeepMind.

Sources and credits

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

  1. Sebastian Borgeaud | alphaXiv alphaxiv.org. Checked 15 October 2026.
  2. Sebastian Borgeaud seb5666.github.io. Checked 15 October 2026.
  3. [2112.04426] Improving language models by retrieving from trillions of tokens arxiv.org. Checked 15 October 2026.
  4. [2112.11446] Scaling Language Models: Methods, Analysis & Insights from Training Gopher arxiv.org. Checked 15 October 2026.
  5. NeurIPS 2022 Awards neurips.cc. Checked 15 October 2026.
  6. [2203.15556] Training Compute-Optimal Large Language Models arxiv.org. Checked 15 October 2026.
  7. An Empirical Analysis of Compute-Optimal Large Language Model Training | ML Anthology mlanthology.org. Checked 15 October 2026.
  8. Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 15 October 2026.
  9. [2312.11805] Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 15 October 2026.
  10. Google launches strike team to improve AI coding models: The Information — Capital Brief capitalbrief.com. Checked 15 October 2026.
  11. Google forms "Strike Team" to improve its coding AI models | heise online heise.de. Checked 15 October 2026.