Building corpora and using corpus data is a core skill, whether the goal is to analyze historical change or diatopic and other kinds of synchronic variation. The lecturers at this summer school introduce a range of cutting-edge techniques and methodologies to help students deal with previously constructed corpora and, crucially, to enable them to build their own. Issues such as the semi-automatic annotation of POS and syntactic structures, the annotation of information structure using the QUD-tree framework, and the BERT machine-learning technique for NLP are covered. Moreover, data types of emerging importance such as sign language and geolocal variation from social-media will be discussed...