The ongoing digitization in the humanities and social sciences (HSS) enables and requires the use of new, particularly quantitative, methods of analysis. Programming skills give researchers the most flexible way of working with data. However, these are not typically taught in HSS degree programs, and existing teaching/learning materials usually require prior technical knowledge. For this reason, the virTUos subproject ExDiMed has developed low-threshold, modular self-learning materials for teaching coding skills in the programming language Python, which are specifically tailored to the needs and requirements of HSS students. The content, in the form of interactive Jupyter Notebooks, is available to learners and teachers as OER and has been suc-cessfully tested and refined at TU Dresden over six semesters in a flipped classroom seminar. With the advent of generative AI, the course has been expanded in an exploratory manner to sensitize students to the constructive use of AI in coding. Our evaluation shows that teaching basic programming skills has become more relevant in the age of AI, as the latter can only act as a catalyst for coding practice if students already have the basics. Our teaching/learning concept is now firmly established in the Digital Humanities master's program at TU Dresden and will con-tinue to be developed after the end of the project.