PeopleofSuperintelligence

Researcher

Alvaro Sanchez-Gonzalez

Alvaro Sanchez-Gonzalez is a machine-learning researcher who applies graph neural networks to physical simulation and weather forecasting. At Imperial College London in 2017, he used machine learning to predict x-ray properties; in 2018 and 2020, he coauthored graph-network simulators, and his 2021 mesh-based simulation paper received an ICLR Outstanding Paper award. In 2023, he contributed to Google DeepMind’s GraphCast, a graph neural weather model that forecasts global conditions up to 10 days ahead; the team received the 2024 MacRobert Award. In 2025, he coauthored research on probabilistic weather ensembles. As of October 2026, he contributes to Google DeepMind’s weather research, including WeatherNext systems.

Timeline

  1. 2017

    Machine learning for X-rays

    In 2017, he leads an Imperial College London study that uses machine learning to predict the properties of x-rays from free-electron lasers. The method could gather useful data up to a thousand times faster.

  2. 2018

    Graph networks as physics engines

    In 2018, he coauthors Graph Networks as Learnable Physics Engines for Inference and Control. Its graph-network model predicts physical-system dynamics, performs system identification and supports gradient-based planning.

  3. 2020

    Graph Network-based Simulators

    In 2020, he coauthors Learning to Simulate Complex Physics with Graph Networks, which introduces Graph Network-based Simulators (GNS). The framework represents particles as graph nodes and computes dynamics via learned message passing to simulate fluids, rigid solids and deformable materials.

  4. 2021

    MeshGraphNets earns an ICLR award

    In 2021, he coauthors Learning Mesh-Based Simulation with Graph Networks, which introduces MeshGraphNets for mesh-based physical simulations. The model passes messages over mesh graphs, adapts mesh resolution during simulation and generalizes to shapes unseen in training; the project receives an ICLR Outstanding Paper award.

  5. 2023

    GraphCast for global forecasts

    In 2023, he contributes to Google DeepMind’s GraphCast research, a graph neural weather model that forecasts global conditions up to 10 days ahead.

  6. 2024

    A MacRobert Award for GraphCast

    In 2024, he is part of the Google DeepMind team whose GraphCast wins the Royal Academy of Engineering’s MacRobert Award for engineering innovation.

  7. 2025

    Probabilistic weather forecasting

    In 2025, he coauthors “Skillful joint probabilistic weather forecasting from marginals,” which introduces FGN for probabilistic weather forecasting. FGN generates ensembles through learned model perturbations and trains to minimize the continuous rank probability score for location-level forecasts.

  8. As of October 2026

    Weather forecasting at Google DeepMind

    As of October 2026, he contributes to Google DeepMind’s AI weather research, including its WeatherNext Cyclones work. He explains how WeatherNext 3 uses direct atmospheric observations to form weather forecasts.

  1. Google had the most people shaping AI every year from 2014 to 2025Includes Google. Our 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. Google has the highest combined h-index among AI organizationsIncludes Google. 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.
  3. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes Google. 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.
  4. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Google. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.

Sources and credits

Photo: We have no published photo of Alvaro Sanchez-Gonzalez yet. Send us a free-licensed photo.

  1. Physicists squeeze extra data from superfast x-ray probes using machine learning imperial.ac.uk. Checked 20 October 2026.
  2. Graph Networks as Learnable Physics Engines for Inference and Control proceedings.mlr.press. Checked 20 October 2026.
  3. Learning to Simulate Complex Physics with Graph Networks proceedings.mlr.press. Checked 20 October 2026.
  4. Learning Mesh-Based Simulation with Graph Networks arxiv.org. Checked 20 October 2026.
  5. Learning mesh-based simulations sites.google.com. Checked 20 October 2026.
  6. GraphCast: AI model for faster and more accurate global weather forecasting deepmind.google. Checked 20 October 2026.
  7. AI weather forecasting tech wins UK’s top engineering award raeng.org.uk. Checked 20 October 2026.
  8. Skillful joint probabilistic weather forecasting from marginals arxiv.org. Checked 20 October 2026.
  9. AI model achieves breakthrough in forecasting cyclones deepmind.google. Checked 20 October 2026.
  10. WeatherNext 3, el modelo con IA que predice fenómenos atmosféricos con mucha más precisión | Tecnología | EL PAÍS elpais.com. Checked 20 October 2026.