PeopleofSuperintelligence

Researcher

John Healy

John Healy is a researcher at the Tutte Institute for Mathematics and Computing in Ottawa whose work centers on unsupervised learning, clustering and dimension reduction. In 2016, as a senior researcher at the institute, he worked on hypergraphs for data analysis and neural networks and spoke on clustering at PyCon 2016. With Leland McInnes, he developed the HDBSCAN clustering library in 2017 and introduced the Uniform Manifold Approximation and Projection (UMAP) dimension reduction method in 2018. He applied manifold learning to single-cell biology in December 2018 and electron microscopy in January 2019, co-led data-science groups at GeekWeek in 2018 and 2019, and co-authored work on vector space embeddings and data maps for cyber defense in July 2024.

Now at
Tutte Institute for Mathematics and Computing

Timeline

  1. 2016

    Senior researcher at the Tutte Institute

    In 2016, he is a senior researcher at the Tutte Institute for Mathematics and Computing, with research in unsupervised learning, hypergraphs for data analysis and neural networks. He presents the talk High Quality High Performance Clustering at PyCon 2016 in Portland.

  2. 2017

    Faster density-based clustering

    In March 2017, the Journal of Open Source Software publishes hdbscan, a software paper he writes with Leland McInnes and Steve Astels. As CSE researchers at the Tutte Institute for Mathematics and Computing, he and McInnes refine and improve the DBSCAN algorithm. In May 2017, they post Accelerated Hierarchical Density Clustering, describing an accelerated HDBSCAN* algorithm with performance comparable to DBSCAN that supports variable-density clusters and eliminates the distance scale parameter.

  3. February 2018–September 2018

    Co-creating UMAP

    In February 2018, he and Leland McInnes post the first UMAP paper to arXiv, introducing Uniform Manifold Approximation and Projection as a manifold learning technique for dimension reduction built on Riemannian geometry and algebraic topology. In September 2018, the Journal of Open Source Software publishes the UMAP software paper, which he writes with McInnes, Nathaniel Saul and Lukas Großberger.

  4. October 2018

    GeekWeek and PyData New York

    In October 2018, he gives an invited talk on Word Embeddings to the Natural Language Processing Group at the National Research Council. At GeekWeek in October 2018, he leads a support group of Tutte Institute and CSE data scientists tasked with giving participants the tools and skills to analyze their data. He also presents the talk HDBSCAN, fast density based clustering, the how and the why at PyData New York City in October 2018.

  5. December 2018–January 2019

    UMAP in biology and materials science

    In December 2018, he and Leland McInnes co-author a Nature Biotechnology paper on dimensionality reduction for visualizing single-cell data with UMAP. Comparing UMAP with five other tools, the study finds that it provides the fastest run times, highest reproducibility and the most meaningful organization of cell clusters. In January 2019, he co-authors an npj Computational Materials paper that uses manifold learning to visualize and explore four-dimensional scanning transmission electron microscopy datasets from single-layer graphene.

  6. November–December 2019

    GeekWeek and PyData Los Angeles

    At GeekWeek in November 2019, he co-leads a team applying Tutte Institute data-science techniques to analyze actors attacking honey-pots on the deep web. In December 2019, he presents the talk MAP all the things at PyData Los Angeles.

  7. February 2023–July 2024

    UMAP history and cyber defense maps

    In February 2023, he and Leland McInnes give the DoEPy talk A History of UMAP As a Python Open Source Project. In July 2024, he and Benoit Hamelin co-author the SciPy presentation Vector space embeddings and data maps for cyber defense, describing a method for embedding data in vector spaces and exploring it through interactive visualization and labeling with data maps.

  8. As of October 2026

    Working at the Tutte Institute

    In March 2026, he co-authors release 0.8.42 of HDBSCAN with Leland McInnes, Steve Astels and others. In 2026, he works at the Tutte Institute for Mathematics and Computing in Ottawa.

  1. Google had the most people shaping AI every year from 2014 to 2025Our 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. The United States employs an estimated 48% of people shaping AIWe estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.
  3. 43% of living people shaping AI work outside their country of birthWe estimate that 43% of living people shaping AI work outside their country of birth (520 of 1,198 people with known birth and employment countries).
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionWe estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

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 John Healy yet. Send us a free-licensed photo.

  1. John Healy | PyCon 2016 in Portland, OR pycon-archive.python.org. Checked 22 October 2026.
  2. hdbscan: Hierarchical density based clustering joss.theoj.org. Checked 22 October 2026.
  3. Accelerated Hierarchical Density Clustering arxiv.org. Checked 22 October 2026.
  4. Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) - Communications Security Establishment Canada cse-cst.gc.ca. Checked 22 October 2026.
  5. UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction arxiv.org. Checked 22 October 2026.
  6. UMAP: Uniform Manifold Approximation and Projection joss.theoj.org. Checked 22 October 2026.
  7. Tutte publications and events - Communications Security Establishment Canada cse-cst.gc.ca. Checked 22 October 2026.
  8. Tutte publications and events - Communications Security Establishment Canada cse-cst.gc.ca. Checked 22 October 2026.
  9. PyData New York City 2018 - John Healy pydata.org. Checked 22 October 2026.
  10. Dimensionality reduction for visualizing single-cell data using UMAP | Nature Biotechnology nature.com. Checked 22 October 2026.
  11. Manifold learning of four-dimensional scanning transmission electron microscopy | npj Computational Materials nature.com. Checked 22 October 2026.
  12. PyData Los Angeles 2019 - Conference Schedule pydata.org. Checked 22 October 2026.
  13. DoEPy - Past Events meetup.doepy.org. Checked 22 October 2026.
  14. Vector space embeddings and data maps for cyber defense | Zenodo zenodo.org. Checked 22 October 2026.
  15. scikit-learn-contrib/hdbscan: release-0.8.42 | Zenodo zenodo.org. Checked 22 October 2026.
  16. jc-healy (John Healy) · GitHub github.com. Checked 22 October 2026.