Purpose: The positive correlation between students' internship and employability is well documented. However, there is no consensus on its relationship with academic performance. Previous studies investigated the Bachelor level: in the UK, the so-called sandwich placement is positively correlated; in the US, the results are mixed. This study aims to expand the scientific literature and focuses on Politecnico di Torino, Italy: the cases of the Industrial and Management Engineering Bachelor (BA) and Master of Science (MS) degree programs are examined. Design/methodology/approach: The academic performance of 2,279 BA students and 2,560 MS students graduated between 2016 and 2021 is examined. Ordinary least square models are employed to test the relationship between grades and internship, controlling for "ex ante" academic scores, gender, age and geographical origin. Findings: The results show a small robust negative correlation between internship and grades, especially at the bachelor level, where potentially a self-selection occurs. At the master level, the negative relationship is significant but very small and might be partially determined by the marks of the exams substituting the internship. Originality/value: This study contributes to the empirical research on the relationship between students' grade and internship and suggests the presence of complementarities in several ways: the Italian context, characterized by an optional internship is examined, both at the BA and MS level. Universities that intend to organize internship as overlapping with lectures and exams should be cautious, especially when considering the BA level students, whose maturity and skills are less developed than MS ones.