Corpus-based interpreting methods are effective for analyzing important phenomena that has been neglected in research (Shlesinger 1998), but little attention has been paid to their possible exploitation in interpreter education and to the benefits of corpus-derived insights for trainee interpreters (Bendazzoli 2010a). The aim of this paper is to describe how Anglintrad, a purpose-specific intermodal Italian-Spanish corpus, is being built and to suggest some preliminary exploitation criteria for interpreter (and translator) training and practice. This paper focuses on the presence of unmodified English loanwords in Italian political speeches (Marzocchi 2007) and their renditions in simultaneous interpreting and written translation into Spanish. The possibility of comparing the same phenomena (unmodified English loanwords) from two different perspectives (interpreting and translation) represents an unprecedented opportunity entailing possible didactic applications to enhance interpreter and translator training and practice.