Course positioning
CS 329S is a project-based ML systems course about the engineering beyond the model: managing data, choosing models, scaling training, monitoring production, and coordinating teams. It also covers privacy, fairness, and security — the real-world issues every system faces.
What it covers
- Data engineering: data management, labeling, versioning, quality
- Model selection: baselines, evaluation, error analysis
- Training and scaling: distributed systems, cost, experiment management
- Deployment: serving, batch/stream, latency and throughput
- Monitoring and iteration: drift, rollback, A/B
- People and society: privacy, fairness, security, teamwork
- Project: design and implement an ML system
Prerequisites
At least one of CS 229/230/231N/224N; familiarity with one framework (PyTorch/TF/JAX).
Assessment & workload
Mainly a project plus assignments/reading. Check the current syllabus for details.
Who it's for
ML engineers moving from model tuning to systems; researchers who want engineering skills; technical product managers.
Study advice
Pair with 'Designing Machine Learning Systems'. The project should include monitoring and rollback, not just a model.
Related courses
Official course description
This project-based course covers the iterative process for designing, developing, and deploying machine learning systems. It focuses on systems that require massive datasets and compute resources, such as large neural networks. Students will learn about data management, data engineering, approaches to model selection, training, scaling, how to continually monitor and deploy changes to ML systems, as well as the human side of ML projects. In the process, students will learn about important issues including privacy, fairness, and security. Pre-requisites: At least one of the following; CS229, CS230, CS231N, CS224N or equivalent. Students should have a good understanding of machine learning algorithms and should be familiar with at least one framework such as TensorFlow, PyTorch, JAX.