The winning proposal of the eScience Center-Lorentz Competition 2025 is:
Beyond Models: Distributed AI Infrastructure as a Scientific Instrument
by
Azza Ahmed, TU Delft
Thomas Abeel, TU Delft
Raymond Oonk, SURF
Andrew Jones, Microsoft
Arnaud Renard, Université de Reims Champagne-Ardenne, MesoNET federation
A roadmap for federated AI
A key enabler of AI is infrastructure that is secure, accessible, scalable, and state-of-the-art. In the Netherlands, such compute capacity remains limited and fragmented despite major initiatives, and power constraints often limit the expansion of local hardware.
This workshop addresses a critical question: How can we share, compute, data, and expertise across institutions safely and effectively? By connecting domain scientists, infrastructure providers, and research software engineers, the organizers aim to move "Beyond Models" to treat distributed AI infrastructure as a unified scientific instrument.
The workshop will bring together a diverse cohort of experts—ranging from CERN and Nikhef to Health-RI and Flower Labs—to explore extreme-scale workloads and AI on regulated, sensitive data.
Workshop with support
As winners of the competition, the organizers will host an intensive, five-day Lorentz Center workshop (lambda location) with a budget of €15,000 for travel, hotel, and catering. Furthermore, the project will receive 460 hours of dedicated support from Research Software Engineers (RSEs) from the Netherlands eScience Center to facilitate the development of a practitioner-oriented decision framework and a "technician’s guide" for federated AI.
Contact
Rena Bakhshi
Programme Officer, Netherlands eScience Center
open-calls@esciencecenter.nl
Tanja Uitbeijerse
Program coordinator, Lorentz Center
competition@lorentzcenter.nl
eScience Center - Lorentz Competition winners
2024
Paradata in 3D Scholarship: Intellectual Transparency and Scholarly Argumentation in Digital Heritage
Isto Huvila, Uppsala University, Sweden
Trilce Navarete, Erasmus University of Rotterdam, Netherlands
Costas Papadopoulos, Maastricht University, Netherlands
Vincent Rossi, The Smithsonian, US
Kira Zumkley, Victoria and Albert Museum, UK
and
Enriching Digital Heritage with Large Language Models (LLMs) and Linked Open Data (LOD)
Arno Bosse, The Royal Netherlands Academy of Arts and Sciences (KNAW), Netherlands
Rossana Damiano, University of Turin, Italy
Leif Isaksen, University of Exeter, UK
Gethin Rees, British Library, UK
Tariq Yousef, University of Southern Denmark, Denmark
2023
A Time Warp in Digital Chemical Discoveries
Titus van Erp, Norwegian University of Science and Technology
Peter Bolhuis, University of Amsterdam
Rosa Bulo, Software for Chemistry & Materials
Roberto Covino, Norwegian University of Science and Technology
and
Modeling the Formation of Respiratory Aerosols
Chantal Darquenne, University of California
Stavros Kassinos, TU Delft
Stefan Hickel, TU Delft
Anna-Carin Olin, University of Gothenburg
Wilbur de Kruijf, Thaerapy BV
