Tech

An AI system to help scientists write expert-level empirical software – Nature

admin01 views

An AI system to help scientists write expert-level empirical software – Nature

  • Article
  • Published:

Nature (2026) Cite this article

We are providing an unedited version of this manuscript to give early access to its findings. Before final publication, the manuscript will undergo further editing. Please note there may be errors present which affect the content, and all legal disclaimers apply.

Subjects

Abstract

The cycle of scientific discovery is frequently bottlenecked by the slow, manual creation of software to support computational experiments1. To address this, we present Empirical Research Assistance (ERA), an AI system that creates expert-level scientific software whose goal is to maximize a quality metric. The system uses a Large Language Model (LLM) and Tree Search (TS)2 to systematically improve the quality metric and intelligently navigate the large space of possible solutions. ERA achieves expert-level results when it explores and integrates complex research ideas from external sources. The effectiveness of tree search is demonstrated across a diverse range of tasks. In bioinformatics, ERA discovered 40 novel methods for single-cell data analysis that outperformed the top human-developed methods on a public leaderboard. In epidemiology, ERA generated 14 models that outperformed the CDC ensemble and all other individual models for forecasting COVID-19 hospitalizations. ERA also produced expert-level software for geospatial analysis, neural activity prediction in zebrafish, and numerical solution of integrals, and a novel rule-based construction for time series forecasting. By devising and implementing novel solutions to diverse tasks, ERA represents a significant step towards accelerating scientific progress.

You have full access to this article via your institution.

Similar content being viewed by others

Author information

Author notes

  1. These authors contributed equally: Eser Aygün, Anastasiya Belyaeva, Gheorghe Comanici, Marc Coram, Hao Cui, Jake Garrison, Renee Johnston, Anton Kast, Cory Y. McLean, Peter Norgaard, Zahra Shamsi, David Smalling, James Thompson, Subhashini Venugopalan, Brian P. Williams

Authors and Affiliations

  1. Google DeepMind, Montréal, Quebec, Canada

    Eser Aygün, Gheorghe Comanici, David Smalling, Anna Bulanova & Shibl Mourad

  2. Google Research, Cambridge, MA, USA

    Anastasiya Belyaeva, Marc Coram, Hao Cui, Renee Johnston, Anton Kast, Cory Y. McLean, Peter Norgaard, Zahra Shamsi, James Thompson, Subhashini Venugopalan, Brian P. Williams, Chujun He, Sarah Martinson, Martyna Plomecka, Lai Wei, Yuchen Zhou, Qian-Ze Zhu, Matthew Abraham, Erica Brand, Jeffrey A. Cardille, Chris Co, Scott Ellsworth, Grace Joseph, Malcolm Kane, Ryan Krueger, Johan Kartiwa, Dan Liebling, Jan-Matthis Lueckmann, Paul Raccuglia, Xuefei Julie Wang, Katherine Chou, James Manyika, Yossi Matias, John C. Platt, Lizzie Dorfman & Michael P. Brenner

  3. Google Platforms and Devices, Mountain View, CA, USA

    Jake Garrison

  4. Massachusetts Institute of Technology, Cambridge, MA, USA

    Chujun He

  5. School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA

    Sarah Martinson, Qian-Ze Zhu, Ryan Krueger & Michael P. Brenner

  6. Google DeepMind, New York, New York, USA

    Martyna Plomecka

  7. Faculty of Agricultural and Environmental Sciences, McGill University, Montréal, Quebec, Canada

    Jeffrey A. Cardille

  8. California Institute of Technology, Pasadena, CA, USA

    Xuefei Julie Wang

Authors

  1. Eser Aygün
  2. Anastasiya Belyaeva
  3. Gheorghe Comanici
  4. Marc Coram
  5. Hao Cui
  6. Jake Garrison
  7. Renee Johnston
  8. Anton Kast
  9. Cory Y. McLean
  10. Peter Norgaard
  11. Zahra Shamsi
  12. David Smalling
  13. James Thompson
  14. Subhashini Venugopalan
  15. Brian P. Williams
  16. Chujun He
  17. Sarah Martinson
  18. Martyna Plomecka
  19. Lai Wei
  20. Yuchen Zhou
  21. Qian-Ze Zhu
  22. Matthew Abraham
  23. Erica Brand
  24. Anna Bulanova
  25. Jeffrey A. Cardille
  26. Chris Co
  27. Scott Ellsworth
  28. Grace Joseph
  29. Malcolm Kane
  30. Ryan Krueger
  31. Johan Kartiwa
  32. Dan Liebling
  33. Jan-Matthis Lueckmann
  34. Paul Raccuglia
  35. Xuefei Julie Wang
  36. Katherine Chou
  37. James Manyika
  38. Yossi Matias
  39. John C. Platt
  40. Lizzie Dorfman
  41. Shibl Mourad
  42. Michael P. Brenner

Corresponding authors

Correspondence to Shibl Mourad or Michael P. Brenner.

Supplementary information

About this article

Cite this article

Aygün, E., Belyaeva, A., Comanici, G. et al. An AI system to help scientists write expert-level empirical software. Nature (2026). https://doi.org/10.1038/s41586-026-10658-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41586-026-10658-6

Related posts

Google announces glasses are back and search is getting an AI makeover – The Guardian

S. ‘Soma’ Somasegar, 1966-2026: Microsoft and Madrona leader was a champion of developers and startups – GeekWire

Android 17’s ‘Continue On’ will let you move tasks between devices, like Apple’s Handoff – 9to5Google