Engineering education is changing - it is evident if you walk through the sessions of any major conference in the field right now. You will find educators rethinking curricula from the ground up, questioning what engineers actually need to know, how they learn to work together, and whether traditional teaching structures are fit for purpose. At SEFI 2025, that conversation was everywhere. But one thread in particular stayed with Dr Jenny Prisutova: the challenge of doing all of this at scale.
My colleague Dr Keith Tarnowski and I attended to present work on the new first-year Mechanical Engineering curriculum that was introduced in 2021. The programme is delivered to over 250 students each year, and the redesign was driven by feedback from students, graduates and industry, telling us that graduates need a better balance of technical and professional skills, and the ability to tackle open-ended engineering problems. The new curriculum places learning in a realistic engineering context from day one, and integrates knowledge, practical, intellectual, and professional skills across two complementary projects, and engineering design modules. The question we set out to answer in our paper was simple: is it working? The answer, mostly, is yes. Students are developing a broader view of engineering, tackling open-ended problems with greater confidence, and engaging more productively with feedback. However, some challenges remain, such as inconsistency in feedback delivery undermining some of the student engagement, and organising substantive projects at this scale putting real pressure on staff, but the direction of travel is encouraging.
What the conference made evident, though, was how differently institutions experience the problem of scale. Talking to colleagues from across Europe, it became clear that a cohort of 250 students feels very different depending on where you are sitting.
Two papers from UCL brought this into sharp focus. The first, by Dr Pilar García Souto and colleagues, examined the impact of different strategies for assigning students to project teams. Their finding was counterintuitive: grouping students so that each team has a balanced spread of prior academic performance did not reliably lead to better group outcomes, but it did tend to produce healthier team dynamics. The study also highlighted a striking disconnect between how students perceive their teamwork and what their output actually shows. This kind of nuanced, evidence-based understanding of team formation is exactly what large programmes need, and it is only possible to develop when you have the scale to run controlled comparisons in the first place. Their use of peer assessment data to evaluate team dynamics has already influenced thinking in our own school, where we are developing a framework for using peer feedback and scores in a summative context.
The second UCL paper, by Dr Fiona Truscott and colleagues, tackled the challenge of running interdisciplinary team projects in what they aptly called "mega classes." Their central argument was that at very large scale, structure is not the enemy of good pedagogy - it is the mechanism through which good pedagogy becomes possible. Regular team meetings, milestone-based assessments, in-person mid-module reviews in place of written reports: these are deliberate design choices that free staff to focus on the interactions that matter most, rather than being consumed by logistics and routine queries. Reading their description of the module lead's role as "curation of staff and student experiences" felt like an articulation of something we have been feeling our way towards ourselves, without quite having the language for it.
What struck me most across these conversations was the degree to which institutions running large programmes have independently arrived at similar conclusions, without necessarily being in dialogue with one another. We are all discovering that scale demands structure, that structure can enable rather than constrain student autonomy, and that assessment design is inseparable from operational feasibility. This convergence is reassuring, but it also points to a gap. Large-cohort engineering education is not simply a scaled-up version of the small-cohort experience - it is a different pedagogical environment, with its own constraints, and its own research questions. It deserves more sustained attention in the literature, and more deliberate exchange between the institutions navigating it.
SEFI 2025 was a reminder of why these gatherings matter. Not
because conferences resolve the hard questions, but because they reveal which
questions other people are asking - and sometimes, that is enough to send you
home with a clearer sense of what you are trying to do.
When citing this work, please use the following
citation:
Prisutova, J (2026). “When Scale Changes Everything: Lessons from
SEFI 2025” Centre
for Engineering Education Blog, The University of Sheffield,
Sheffield, UK. May 2026.