"Internationalisation" has become the new buzzword for universities around the world, with jointly offered degrees as well as smaller-scale exchanges for students. Despite this rapid expansion of international campuses and programmes, and the increasing acceptance and encouragement of international experiences for [post]graduate students, little comprehensive evaluative work has been done to assess their efficacy on a broad scale and to determine what types and models of international work can be most effective. The lack of reliable and comprehensive data is especially problematic for science and engineering fields, where academic staff anxieties about forming students into competent scientists often collide with enthusiasm for encouraging international collaborations. Questions of exactly what makes a competent, or excellent, scientist, and what may benefit the scientific domain, do not have easy--or agreed-upon--answers. This article assesses the current state of internationalisation and international experiences, focusing in particular on science and engineering fields. It discusses initial results from a workshop, sponsored by the US National Science Foundation and organised by the Center for Innovation and Research in Graduate Education at the University of Washington, to develop an interdisciplinary research agenda aimed at launching and coordinating empirically driven research on international graduate education. It concludes by identifying areas for future research. (Contains 1 figure.)