This unit guides students through configuring the AWS CLI and preparing a development environment to use Amazon Bedrock, covering both textual and multimodal model interactions. Students will learn inference strategies (zero-, one-, few-shot), prompt design, access control, and how to integrate core AWS services into an end-to-end generative AI solution.
Learning Objectives
- Analyze the differences, strengths, and appropriate use cases for zero-shot, one-shot, and few-shot inference strategies and justify selection for specific AI tasks
- Apply AWS CLI and Bedrock environment configuration steps to authenticate, set regions and profiles, and establish a reproducible local/cloud development setup
- Demonstrate creation, testing, and iterative refinement of prompts for textual and multimodal foundational models via API calls, including evaluation of output quality and failure modes
- Evaluate and implement access control and AWS service integrations (e.g., IAM policies, S3, Lambda, and other ecosystem services) to design and document a secure end-to-end generative AI solution
Curriculum
- 1 Section
- 9 Lessons
- 30 hours
Expand all sectionsCollapse all sections
- Unit 2 — Bedrock Foundations and AWS Integration9
- 1.1Next Steps: Configuring the AWS CLI20 mins
- 1.2Tech Setup for Using Generative AI on AWS20 mins
- 1.3Learning About Zero-, One- and Few-Shot Inference20 mins
- 1.4Using Textual Models Through the API20 mins
- 1.5Creating Prompts for Multimodal Foundational Models20 mins
- 1.6Integrating Access Control Management20 mins
- 1.7Leveraging the AWS Ecosystem in an End-to-End Solution20 mins
- 1.8Amazon Bedrock Capabilities20 mins
- 1.9Configuring Your Bedrock Environment20 mins
Instructor
Free
Leave a Reply