Systems for Large-Scale ML

3.0 credits · Graduate

Systems for Large-scale ML is a new advanced course at Cornell University designed to equip students with hands-on expertise in designing and operating scalable machine learning systems. With the rise in popularity of large ML models like GPT, LLaMA, and DeepSeek, tackling systems-level challenges of distributing training and inference workloads on multi-accelerator hardware while ensuring fault tolerance has become a crucial skill for both graduate and undergraduate students in computer science. The course will teach students to tackle systems challenges in both training and inferring from large-scale ML models. We will combine theory and hands-on teaching through lectures, assignments, and projects.