Minisymposium

CodeRefinery is a community-driven initiative that equips researchers and students across disciplines with essential research software engineering skills to develop sustainable, open, and FAIR (Findable, Accessible, Interoperable, Reusable) software. Funded in-kind by multiple Nordic organizations and the Nordic e-Infrastructure Collaboration (NeiC), CodeRefinery addresses critical gaps in traditional academic education by providing hands-on training in version control, reproducible research, collaborative development, and efficient coding practices that foster Open Science. Our workshops are designed to be accessible and scalable, featuring live-streamed and recorded sessions with interactive participation via collaborative documents. All training materials (available at https://coderefinery.org/lessons/) are open-source, ensuring reusability and adaptation by the broader research community. By working with local partners, CodeRefinery fosters an interactive and collaborative learning environment, overcoming challenges associated with large-scale and online teaching. This community-driven approach has enabled the joint organization of specialized workshops, such as high-performance computing (HPC) kick-off and “Tools and Techniques for HPC”, addressing a wide range of computational research needs. Through these efforts, we contribute to a growing ecosystem of training initiatives that empower researchers to build sustainable and reusable research software.

Standardization is one cornerstone in Training unification across Europe, and the HPC Certification Forum is pursuing this task using a Skill tree. This tree starts with Skill categories such as HPC Knowledge or Software development and branches out into smaller and smaller nodes until a single skill is reached. A skill in this case is a single unit of learning containing knowledge that can be learned in 1 to 4 hours with lecture or reading and self study time. The core of each skill are universally applicable learning objectives formulated using Bloom's Taxonomy. Furthermore, an examination framework allows training providers as well as participants to certify the knowledge acquired on a skill to skill basis. This framework will not be used for the new EuroHPC Training platform under hpc-portal.eu for categorizing training events. This is done under the HPC SPECTRA project and should be rolled out in Spring of 2025. Additionally, the skill framework will be used to create learning pathways and baselines in collaboration with the CASTIEL2 project.

Research Software Engineers (RSEs) combine professional software engineering expertise with a good understanding of research practices, ensuring sustainable and reproducible research outputs. In High-Performance Computing (HPC) RSEs play a critical role in the research pipeline, but there are no established routes for becoming an RSE and developing required skills. Additionally, the increasingly diverse backgrounds of RSE and HPC professionals make training provision and professional development more difficult. Many professionals are forced to discover, develop and progress their skills on the job, which can be challenging and time consuming. Finding the right content can be difficult, especially outside of the higher education context. To ensure sustainable growth of research communities, it is necessary to expand and improve existing HPC and RSE education and training programmes to mirror technological changes, and provide people in research software roles with the right set of skills. It is also necessary to make the HPC and RSE training ecosystem more FAIR (Findable, Accessible, Interoperable and Reusable) and to explore new approaches to skills development in both formal and informal settings. This presentation will briefly describe the work done within the UNIVERSE-HPC project (https://www.universe-hpc.ac.uk/), and plans for the follow-up projects - DRIFT and CHARTED.

This session we will discuss how developer and training communities can collaborate to ensure that the training offered aligns with the needs of the research and developer communities and how researchers and developers can contribute to make training initiatives more targeted to their needs We will address the challenges of onboarding a researcher into the research software engineering world and how to fill knowledge gaps in the current HPC and sustainable software development ecosystem. The discussion is moderated by the organizer. All speakers will be involved in the discussion round and we welcome researchers of all scientific domains, and representatives of all training and scientific software developer communities to join us.