Speaker
Description
What happens when a three-person development team moves from being embedded in a domain research group to operating more as a dedicated RSE group within the same department? Four months ago, our group at SINTEF Community made exactly that shift while continuing to support the same applied research environment in the built environment sector. The domain, collaborators, and technical challenges remained largely the same, but the organisational relationship changed. That transition became a useful lens for examining questions that RSE discussions often treat at a general level.
What do you lose when you step back from full embeddedness, and what do you gain? When we were part of the domain group, proximity shaped everything: we understood research needs quickly, caught problems early, and moved fast. At the same time, we struggled to protect time for reusable infrastructure, longer-term engineering decisions, and work beyond the next urgent prototype. The new structure creates more room for deliberate software engineering, while raising new questions about how to stay close enough to the domain to build the right things.
In this lightning talk, I reflect on four tensions that this transition made visible: domain proximity versus engineering discipline; when a prototype has earned real software engineering investment; how a three-person team maintains breadth across ontology engineering, AI pipelines, and live sensor infrastructure; and what research software sustainability means when your users are also your colleagues.
The talk is an open reflection for RSEs thinking about where their group sits, or should sit, relative to the researchers they support.