Machine Learning Operations (COMP6984001)

Graduate course at BINUS University on the lifecycle of production ML systems, based on Gift and Deza’s Practical MLOps (O’Reilly).

Sessions

• Sessions 1-3: MLOps Foundations — introduction to MLOps and the ML lifecycle, MLOps solutions, model development and implementation
• Sessions 4-5: Object-oriented Programming for ML — software engineering foundations for production ML code
• Sessions 6-8: Pipelines and Deployment — machine learning pipelines, deploying your first model, deploying at scale
• Sessions 9-10: Model Deployment with FastAPI — serving models through production APIs
• Session 11: Testing and Securing ML Solutions
• Sessions 12-13: Model Monitoring — performance, drift detection, and alerting

Assessment

Applied project building and operating a production ML service, plus written components.

Resources

• Textbook: Gift, Deza, and others. Practical MLOps: Operationalizing Machine Learning Models (O’Reilly)
• Tools: MLflow, Docker, Kubernetes, FastAPI, Prometheus, Grafana

The outline and assessment design are summarized here; full session materials are shared with enrolled students only.