This course introduces students to the principles and practice of data science in economics and finance. The goal is not only to learn technical tools, but to understand how empirical researchers design, evaluate, and interpret data-driven analyses. The course is structured around the data science pipeline: formulating questions, constructing datasets, representing data, estimating models, evaluating predictions, and translating results into decisions. Along the way we develop a framework for distinguishing prediction from causal inference, understanding heterogeneity in economic data, and evaluating models critically rather than treating them as black boxes. Students will learn how to work with structured and unstructured data, including text, and will gain exposure to modern machine learning methods. The course also introduces large language models and their role in modern empirical workflows, including how they can be used for data construction, labeling, and analysis. The course emphasizes verification, reproducibility, and critical evaluation of model outputs. Students are expected to attend all classes, actively participate in discussions, and complete two in-class exams in addition to problem sets.  Course open only to juniors and seniors.  This is not a writing intensive course.