Research facilities face a unique software development challenge: delivering reliable and performant software systems on the lifetime of an experiment set up, while serving a diverse and demanding user community. Agile methodologies offer an interesting framework for this context, although its adoption across large-scale scientific infrastructure remains uneven.
This full-day workshop brings together software engineers, project managers, and scientists from user facilities to explore how Agile principles can be applied effectively in this environment — and how the emergence of Agentic AI is beginning to reshape that landscape.
Program overview
The morning session will present concrete experience from facilities that have successfully adopted or adopting Agile practices. Speakers will share what worked, what failed, and the adaptation of Agile frameworks to the scientific software environment.
The afternoon session turns to a new perspective. Agentic AI systems, capable of autonomous planning, code generation, and multitask execution, are no longer a distant future. Aiming as open discussion, this session is focused on their impacts for research facility through an Agile lens: how AI agents will transform the relationship between developers and users; and what this means for the education and mentorship of the next generation of research software engineers.
Contacts: Vicent Hardion (vincent.hardion@ess.eu) and Lukasz Burdzanowski (lukasz.burdzanowski@cern.ch)
