RSE across the NFDI - Connecting Infrastructure and Community through FAIR4RS Practice

Research Software Engineers (RSEs) have become an essential driving force behind the sustainable creation, curation, and FAIRification of research data. Within the German National Research Data Infrastructure (NFDI), this role is gaining increasing formal recognition – not only as software developers but as integral contributors to infrastructure design, metadata standards, and cross-consortia interoperability.

The Working Group “Research Software Engineering” (WG RSE) of the NFDI’s Section Common Infrastructures brings together RSEs from interdisciplinary NFDI consortia. It develops a shared understanding of RSE roles, best practices, and common frameworks for research software within data infrastructures. The group’s conceptual foundation builds on three pillars that position RSEs as:

1. Creators of Research Software as Research Artefacts,
2. Providers of FAIRification Services,
3. and Contributors to Infrastructure Services such as Base4NFDI’s basic services (nfdi.software, Jupyter4NFDI, TS4NFDI).

Across these pillars, the WG coordinates five cross-cutting tasks (CCTs) that define practical RSE activities in research data infrastructures:

(A) creating metadata and applying FAIR4RS principles;
(B) implementing software quality and sustainability criteria;
(C) contributing to archiving and infrastructure integration;
(D) making tools FAIR and visible through marketplaces; and
(E) creating shared guidelines and training opportunities.

This structured approach serves as both a coordination framework and a point of connection between RSEs in the NFDI and the wider international RSE community. The paper presents this model and invites discussion about its transferability beyond the NFDI context – for instance, how similar structures could support RSE communities in other national or institutional frameworks.

By showcasing examples such as the SPARQLing Unicorn QGIS Plugin, the Jupyter Python Minions, and the Academic Meta Tool (all from the archaeological domain), the paper visualises how RSE work in NFDI4Objects, NFDI4ING, or NFDI4Energy bridges domain boundaries and fosters reproducible, FAIR, and open research workflows. These examples demonstrate that the technical and social aspects of Research Software Engineering are deeply intertwined: developing FAIR software is not only about compliance, but about creating communities of practice.

This paper serves as an invitation to dialogue: How can the broader RSE community and the NFDI ecosystem learn from one another? How can international RSE initiatives and infrastructures collaborate to strengthen FAIR4RS implementation and Research Software sustainability across disciplinary borders?