This course is designed to equip students with a basic to advanced understanding of the concepts and practices of Machine Learning Operations (MLOps) for deploying AI systems in real-world environments. In this course, students will learn the machine learning lifecycle as a whole, from data ingestion, preprocessing, and versioning to model training, evaluation, and experiment tracking. The discussion continues with reproducibility and collaboration practices, including version control for code, data, and models, which shows how disciplined engineering workflows turn experimental models into reliable, maintainable systems.
Next, students will explore the automation of machine learning workflows, with a focus on continuous integration and continuous delivery (CI/CD) pipelines, automated testing, and pipeline orchestration. This course also covers model deployment strategies, including containerization, serving models through REST APIs, and batch versus real-time inference, as well as important issues such as performance monitoring, data drift, model drift, and retraining strategies. In the final sessions, students are introduced to the evaluation and selection of MLOps tools and cloud platforms, along with the governance, security, and ethical considerations of operating AI systems in production.
Through a combination of conceptual foundations and hands-on implementation, this course provides comprehensive skills for designing, building, deploying, and maintaining production-ready machine learning systems. By the end of the course, participants are expected to be able to design and implement end-to-end MLOps pipelines, apply CI/CD principles to machine learning projects, deploy and monitor models in production, and justify their choice of MLOps tools and platforms based on scalability, cost, and use case.
