This course aims to give a comprehensive introduction to causal inference. Targeted at Data Science masters students in their second semester, the course introduces students to the potential outcomes and directed acyclic graph frameworks for formalizing the assumptions necessary to support causal claims. Emphasis is on the general logic of non-paramteric causal identification and design-based inference for causal estimands. By the end of the course, students should be able to formulate research designs constrained by the set of assumptions plausible for a specific empirical application. All classes divide time between theory and hands-on application in R and Python.