Moscow Institute of Physics and Technology
New Economic School
Carolina ParadaVP and Head of Robotics · Google DeepMindResearcher↗
Researcher · Engineer
Elena Buchatskaya
Google DeepMindSenior Staff Research Engineer
Buchatskaya co-authored BYOL, which showed that vision systems can learn useful representations from images without labels or contrasting negative examples.[1] She also contributed to Chinchilla, which demonstrated that language models often improve more by training on additional data than by simply adding parameters.[2]
Career & education
Current roleGoogle DeepMind2025-presentSenior Staff Research Engineer
Google DeepMind2025-presentPreviouslyGoogle DeepMind2018-2025Research Engineer -> Staff Research Engineer
Goldman Sachs2015-2018
Google DeepMind2018-2025Goldman Sachs2015-2018Work and education history is primarily focused on AI-relevant roles and may not be comprehensive.
Citation Trend
Total citations to scholarly work, including research papers, technical reports, books and theses, as recorded by Google Scholar. Each point is a dated snapshot.
citations53,796As of Oct 22, '26
citations+11,301Since last snapshot, Jun 15, '26
Selected contributions
Training Compute-Optimal Large Language Models
With Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Trevor Cai, Eliza Rutherford, Diego de las Casas, Lisa Anne Hendricks and others ·Chinchilla showed how smaller models trained on more data can better use a fixed computing budget.
References
- Bootstrap Your Own Latentarxiv.org
Original BYOL paper credits Buchatskaya and demonstrates learning without labels or negative pairs.
- Training Compute-Optimal Large Language Modelsarxiv.org
Original Chinchilla paper credits Buchatskaya and studies optimal balance of model size and training data.
Based on public sources. If accuracy is important, verify details using the links or references.
For corrections, updates, or requests to remove information, contact us.
Reviewed August 8, 2026 · Checked for accuracy and new information.
