PeopleofSuperintelligence

Researcher

Mario Geiger

Mario Geiger is a physicist whose AI research builds neural networks that encode geometric symmetries. He coauthored Spherical CNNs, which won the ICLR 2018 Best Paper Award, and developed e3nn for E(3)-equivariant learning. He also coauthored NequIP for molecular dynamics and Proteina for protein structure generation. As of October 2026, he is a senior research scientist with NVIDIA's BioNeMo team.

Now at
NVIDIA
Badges

Timeline

  1. 2018

    Spherical CNNs win an ICLR award

    With Taco S. Cohen, Jonas Koehler and Max Welling, he coauthors Spherical CNNs in 2018, introducing rotation-equivariant cross-correlation for spherical data. The paper wins the ICLR 2018 Best Paper Award.

  2. 2021

    A PhD in physics

    He completes his PhD in EPFL's Physics of Complex Systems Laboratory in 2021 under Matthieu Wyart. His research there examines neural-network dynamics and the theory of equivariant networks.

  3. 2021

    e3nn and NequIP

    In 2021, he is the main developer of e3nn, a library for equivariant neural networks. With Simon Batzner and collaborators, he coauthors NequIP, an E(3)-equivariant graph neural network for learning interatomic potentials in molecular dynamics simulations.

  4. 2022

    A postdoc at MIT

    In 2022, he is a postdoctoral researcher at MIT with Prof. Smidt and gives a talk on equivariant neural networks.

  5. 2022

    A formal e3nn framework

    He and Tess Smidt present e3nn in a 2022 paper as a generalized framework for E(3)-equivariant trainable functions on 3D geometry and geometric tensors.

  6. November 2024

    NVIDIA cuEquivariance

    On November 18, 2024, he coauthors NVIDIA's cuEquivariance launch post, describing a CUDA math library that accelerates equivariant neural networks for drug discovery and materials science.

  7. 2025

    Proteina for protein design

    He coauthors Proteina in 2025, a flow-based protein-backbone generator that uses hierarchical fold-class labels to guide generation.

  8. As of October 2026

    NVIDIA BioNeMo research

    As of October 2026, he is a senior research scientist with NVIDIA's BioNeMo team.

Career

  1. NowNVIDIASenior Research Scientist
  2. From 2022Massachusetts Institute of TechnologyPostdoctoral Researcher

Education

  1. 2021EPFLPhD

From the biography.

  1. 16% of AI career moves in the 2020s went from a university to a companyIncludes Mario Geiger. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  2. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Mario Geiger. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  3. Google has the highest combined h-index among AI organizationsIncludes NVIDIA. 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.
  4. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes NVIDIA. 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.

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 Mario Geiger yet. Send us a free-licensed photo.

  1. Spherical CNNs arxiv.org. Checked 21 October 2026.
  2. 10 Cutting Edge Research Papers In Computer Vision & Image Generation topbots.com. Checked 21 October 2026.
  3. Matthieu Wyart people.epfl.ch. Checked 21 October 2026.
  4. MRS 2021 Fall Meeting Tutorial: Symmetry-Aware Neural Networks for the Materials Science with e3nn e3nn.org. Checked 21 October 2026.
  5. E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials arxiv.org. Checked 21 October 2026.
  6. Group Theory for Machine Learning slides.com. Checked 21 October 2026.
  7. e3nn: Euclidean Neural Networks arxiv.org. Checked 21 October 2026.
  8. Accelerate Drug and Material Discovery with New Math Library NVIDIA cuEquivariance developer.nvidia.com. Checked 21 October 2026.
  9. Proteina: Scaling Flow-based Protein Structure Generative Models arxiv.org. Checked 21 October 2026.
  10. Author: Mario Geiger | NVIDIA Technical Blog developer.nvidia.com. Checked 21 October 2026.