AI Model Deployment and Governance

This unit explores how foundation models and machine learning systems are designed, evaluated, and operationalized using AWS tools while emphasizing responsible, secure, and compliant practices. Students will deepen technical knowledge of model internals (tokens, embeddings, transformers), learn MLOps pipelines and AWS AI services (including Amazon Bedrock and SageMaker), and assess ethical, security, and suitability considerations for real-world deployments.

Learning Objectives

• Analyze the relationships among AI, ML, and deep learning and explain the transformer architecture, foundation models, tokens, chunking, and embeddings to justify modeling choices
• Apply AWS AI services (such as Amazon Bedrock and SageMaker) and construct a basic MLOps pipeline to demonstrate model customization, inference tuning, and knowledge-base integration
• Demonstrate model evaluation and monitoring techniques, including the effects of inference parameters on outputs, and implement basic model customization and rollout strategies
• Evaluate responsible AI, security, governance, and compliance requirements to identify unsuitable use cases and recommend mitigation strategies for ethical and secure deployments