In multidisciplinary technology-based engineering diploma programmes in South Africa, the curriculum is often structured into distinctly theoretical and practical components, each of which is taught and assessed at different stages by different disciplinary or technical specialists. This separation does not necessarily reflect the complexity of such emerging regions, nor allow for the opportunity to assess multidisciplinary competence in the context of real world practice. Although the Exit Level Outcomes, endorsed by the Engineering Council of South Africa, are intended to provide a holistic framework of achievement in engineering qualifications, it is evident that these outcomes mean different things to the various stakeholders involved in curriculum design, delivery and evaluation. The moment of final academic assessment presents a number of challenges. Who is in a position to assess whether or not a candidate has successfully demonstrated the required level of competence? Legitimation Code Theory offers a lens through which to consider the nature of and relationship between the epistemic and social aspects of the assessment of complex performance, particularly in the South African socio-historical context. This paper presents the analysis of a single engineering assessment case study in which the knowledge and knower values that emerged among a group of assessors are interrogated.'The findings suggest that in the absence of specific epistemic expertise, the default assessment position relies on knower attributes. This may have implications for the assumption in science-based professions that what you know matters more than who you are.