PeopleofSuperintelligence

Researcher

Varun Kanade

Varun Kanade is a computer scientist who studies machine learning foundations and how Transformers learn from examples. He earned a BTech at the Indian Institute of Technology in Mumbai in 2006, studied at Georgia Tech and Harvard, and joined UC Berkeley as a Simons Postdoctoral Fellow in 2012. His work ranges from hierarchical clustering to fairness in prediction, including how global calibration strengthens multiaccuracy. He received a Turing Fellowship in 2024 and is an Associate Professor of Computer Science at Oxford and a Chen StarVerse Fellow at Lady Margaret Hall.

Now at
Lady Margaret Hall, University of Oxford and European Laboratory for Learning and Intelligent Systems

Timeline

  1. 2006

    A BTech in Mumbai

    He earns a BTech at the Indian Institute of Technology in Mumbai, India in 2006.

  2. 2009–2012

    Georgia Tech and Harvard

    Before Harvard, he studies at Georgia Tech with Adam Kalai. From fall 2009 to summer 2012, he is a PhD student at Harvard, with Leslie Valiant as his adviser.

  3. 2012

    A Simons postdoctoral fellowship

    He joins UC Berkeley as a Simons Postdoctoral Fellow in fall 2012.

  4. 2018

    Hierarchical clustering objectives

    At SODA 2018, he coauthors a paper that uses an axiomatic approach to define objectives for similarity and dissimilarity-based hierarchical clustering.

  5. 2024

    Learning inside Transformers

    At ICLR 2024, he coauthors a study of in-context learning over Boolean functions. The study finds that Transformers nearly match the optimal learning algorithm on simpler tasks, with weaker performance on complex ones.

  6. 2024

    An Alan Turing Fellowship

    The Alan Turing Institute awards him a Turing Fellowship in 2024. His work covers the foundations of machine learning and the algorithmic properties of foundation models.

  7. 2025

    Calibration and multiaccuracy

    In 2025, he coauthors a paper on multiaccuracy and multicalibration, two multigroup fairness notions for prediction. It finds that global calibration strengthens multiaccuracy's power.

  8. As of October 2026

    Associate Professor at Oxford

    As of October 2026, he is an Associate Professor of Computer Science at Oxford and a Chen StarVerse Fellow in Computer Science and a Governing Body Fellow at Lady Margaret Hall. ELLIS lists him as a Scholar in the Oxford unit.

Career

  1. NowLady Margaret Hall, University of OxfordChen StarVerse Fellow in Computer Science and Governing Body Fellow
  2. NowEuropean Laboratory for Learning and Intelligent SystemsScholar in the Oxford unit
  3. NowUniversity of OxfordAssociate Professor of Computer Science
  4. From 2012University of California, BerkeleySimons Postdoctoral Fellow

Education

  1. 2012Harvard UniversityPhD Student
  2. 2006IIT BombayBTech
  3. Georgia Institute of TechnologyGraduate Student

From the biography.

Checked career sources: simons.berkeley.edu, lmh.ox.ac.uk, cs.ox.ac.uk.

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 Varun Kanade yet. Send us a free-licensed photo.

  1. Varun Kanade simons.berkeley.edu. Checked 26 October 2026.
  2. Hierarchical Clustering: Objective Functions and Algorithms arxiv.org. Checked 26 October 2026.
  3. Research cs.ox.ac.uk. Checked 26 October 2026.
  4. Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete Functions proceedings.iclr.cc. Checked 26 October 2026.
  5. Varun Kanade awarded prestigious Turing Fellowship cs.ox.ac.uk. Checked 26 October 2026.
  6. How Global Calibration Strengthens Multiaccuracy machinelearning.apple.com. Checked 26 October 2026.
  7. Varun Kanade | Lady Margaret Hall lmh.ox.ac.uk. Checked 26 October 2026.
  8. Varun Kanade | Department of Computer Science cs.ox.ac.uk. Checked 26 October 2026.
  9. Varun Kanade | European Laboratory for Learning and Intelligent Systems ellis.eu. Checked 26 October 2026.