Current courses — 2026
Harvard now lists these courses as COMPSCI 1090A and COMPSCI 1090B (formerly CS109A and CS109B).
- COMPSCI 1090A: Data Science 1 — Introduction to Data Science, Fall 2026
- COMPSCI 1090B: Data Science 2 — Advanced Topics in Data Science, Spring 2026
The official Harvard listings provide course information and links to Canvas. Course access may require a Harvard login. The historical course materials below remain available for independent study.
Course overview and archives
Harvard's CS109A course is an introductory course in data science, designed for students with some prior programming experience. The course covers a broad range of topics in data science, including data cleaning, visualization, analysis, and machine learning. Students learn to use tools such as Python to work with data and build statistical models. The course also includes a significant group project component, where students work in teams to apply their skills to a real-world data science problem. Overall, the course aims to provide students with a solid foundation in data science and prepare them for further study in the field.
CS109B is a more advanced course in data science at Harvard, focused on topics in deep learning and probabilistic machine learning. The course covers techniques for building and training deep neural networks, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), as well as more recent developments in the field such as transformers and language models like BERT and GPT, GANs, Style Gans. Additionally, the course explores unsupervised learning techniques, such as clustering, and covers probabilistic machine learning in a Bayesian framework. Overall, the course is designed to provide students with a deep understanding of advanced techniques in data science and prepare them for research or careers in the field.
From 2013 to 2015, it was taught as a one-semester course by Hanspeter Pfister and Joe Blitzstein. In 2016 the course expanded to two semesters, CS109A and CS109B, and has been taught by Pavlos Protopapas from SEAS and Natesh Pillai, Kevin Rader, and Mark Glickman from Statistics.
The diagram below extends the course history through 2026 and includes 2027 as a future year. Historical course links are preserved; outlined year markers do not imply a public archive or a confirmed future offering.
CS109B is a more advanced course in data science at Harvard, focused on topics in deep learning and probabilistic machine learning. The course covers techniques for building and training deep neural networks, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), as well as more recent developments in the field such as transformers and language models like BERT and GPT, GANs, Style Gans. Additionally, the course explores unsupervised learning techniques, such as clustering, and covers probabilistic machine learning in a Bayesian framework. Overall, the course is designed to provide students with a deep understanding of advanced techniques in data science and prepare them for research or careers in the field.
From 2013 to 2015, it was taught as a one-semester course by Hanspeter Pfister and Joe Blitzstein. In 2016 the course expanded to two semesters, CS109A and CS109B, and has been taught by Pavlos Protopapas from SEAS and Natesh Pillai, Kevin Rader, and Mark Glickman from Statistics.
The diagram below extends the course history through 2026 and includes 2027 as a future year. Historical course links are preserved; outlined year markers do not imply a public archive or a confirmed future offering.
How to Take this Course
Independent study: Browse the historical course links above, including CS109A 2021 and CS109B 2022.
HarvardX online courses
Explore all three courses on Harvard’s official course pages for current enrollment and certificate options:
- Introduction to Data Science with Python
- Machine Learning and AI with Python
- Introduction to Neural Networks and Deep Learning with Python
Harvard Extension School
Consult the course catalog for current offerings, registration requirements, and credit information.
Last reviewed November 14, 2026.