Offered: Fall 2025 (current)
Overview of ML Systems, software engineering principles, data engineering fundamentals; Data Management and Feature Engineering: data collection and preprocessing, feature engineering, feature stores, data validation and monitoring; Model Development and Training: model selection and evaluation, distributed training, model versioning and reproducibility, MLOps for model training; Model Deployment and Serving: deployment strategies, model serving infrastructure, performance optimization, model monitoring and logging; Scalability and Reliability: scalable system design, real-time ML systems, handling data and model drift, system reliability and resilience; Advanced Topics and Case Studies: MLOps best practices, edge AI and IoT applications, responsible AI in production, case studies, ML system security.
The core objectives of this course are to:
To provide a comprehensive overview of the architecture and components of ML systems
To develop proficiency in data collection, preprocessing, feature engineering, and feature store management
To gain expertise in model selection, distributed training, model versioning, and MLOps for training
To learn deployment strategies, model serving infrastructure, and performance optimization techniques
To understand principles of scalable system design, real-time ML systems, and handling data and model drift
To explore best practices, edge AI, responsible AI, and security in ML systems
To understand how standard software engineering principles relate to ML systems.
1. To Be Added
| # | Description | Weight | Edit |
|---|
