Production AI on AWS

Description

This three-day, instructor-led course equips experienced engineers and architects with the skills to design, build, and operationalize production-grade AI and generative AI solutions on AWS. Participants work hands-on with Amazon SageMaker Unified Studio across the end-to-end ML lifecycle — from data preparation and model training through deployment and monitoring — and with the Amazon Bedrock suite to deploy, customize, and govern foundation models for enterprise use cases. The course emphasizes real-world architecture patterns that meet scalability, security, and cost requirements, and prepares participants to provide design and solution leadership for AI initiatives. Labs are goal-oriented: participants receive objectives and success criteria, not step-by-step instructions, reflecting the advanced experience level of the cohort. A CloudFormation foundation stack provisions the shared environment (SageMaker Unified Studio domain, S3 data, Bedrock Knowledge Base, IAM roles, Lambda stubs) before the course begins, enabling each lab to run independently without dependencies on other labs. Participants receive 7 days of lab environment access — the 3-day course plus 4 additional days for independent practice and lab completion. NOTE: This is a skills-based course, not aligned to a specific AWS certification exam. It is positioned on the AWS AI/ML learning path after the AIF-C01 (AI Practitioner) and MLA-C01 (ML Engineer) courses. Participants with backgrounds in Azure or GCP will find cross-cloud comparisons woven into the instruction.

Who This Course Is For

• Data engineers with 8+ years of experience building data pipelines and ML infrastructure across AWS, Azure, or GCP
• Solutions architects and cloud architects designing AI/ML platforms for enterprise workloads
• Senior engineers and engineering leads responsible for production AI systems and MLOps
• Technical leads translating business requirements into AI architecture decisions
• Platform engineers building shared AI/ML infrastructure for development teams

What You Will Learn

• Evaluate SageMaker Unified Studio as an integrated AI development platform and configure it for enterprise data and ML workflows.
• Build and manage end-to-end ML workflows using SageMaker Unified Studio, including data preparation, model training, evaluation, deployment, and monitoring.
• Evaluate, deploy, and customize foundation models using Amazon Bedrock for enterprise generative AI use cases.
• Design and build agentic AI applications using Bedrock Agents, Flows, and Knowledge Bases for complex enterprise workflows.
• Implement responsible AI practices and governance frameworks for AI solutions on AWS using Bedrock Guardrails, SageMaker Clarify, and compliance tooling.
• Design production-ready AI architectures using real-world patterns that meet scalability, security, and cost requirements, and provide design and solution leadership for AI initiatives.

Prerequisites

• 3+ years of hands-on experience with AWS services (IAM, S3, VPC, Lambda, CloudFormation or CDK)
• Familiarity with machine learning concepts: training, inference, model evaluation, and deployment
• Experience with at least one ML framework (TensorFlow, PyTorch, Scikit-learn, or equivalent)
• Working knowledge of Python and SQL
• Comfort with the AWS Management Console and AWS CLI
• An AWS account with administrative access for hands-on labs
• Basic understanding of generative AI concepts (foundation models, tokens, embeddings, RAG) — equivalent to AIF-C01 level knowledge

Delivery Options

• Live, instructor-led