Developing Generative AI Applications on AWS
Please note that this is not a prep course for any AWS certification!
This document contains links mentioned in the “Developing Generative AI Applications on AWS” course.
Module 0: Introduction
Full name: Developing Generative AI Applications on AWS
Product code: 300-MLDGAI-21
These links are based on the course version 2.1.16.
Online and classroom course options: Developing Generative AI Applications on AWS on AWS Skill Builder
The official course material: eVantage Bookshelf (VitalSource)
Lab environment: AWS Builder Labs
Additional material: AWS Skill Builder
Module 1: Exploring Components of Generative AI Applications on AWS
Websites
- What Is Artificial Intelligence (AI)?
- Generative AI on AWS
- What Is Generative AI?
- What are Foundation Models?
- What are Transformers in Artificial Intelligence?
- Generative AI use cases
- AI Services
- What is Amazon SageMaker AI?
- Welcome to a New Era of Building in the Cloud with Generative AI on AWS
- Supported foundation models in Amazon Bedrock
Research papers
Module 2: Programming with Amazon Bedrock
Websites
Module 3: Applying Prompt Engineering for Developers
Websites
- Prompt engineering concepts in the Amazon Bedrock User Guide
- Anthropic Claude prompt engineering guide
- Prompt templates and examples for Amazon Bedrock text models
Research papers
- ReAct: Synergizing Reasoning and Acting in Language Models
- ART: Automatic multi-step reasoning and tool-use for large language models
Module 4: Using Amazon Bedrock APIs in Common Architectures
Websites
- Layer 4: Repeatable application patterns for common generative AI use cases
- How inference works in Amazon Bedrock
- Amazon Bedrock Pricing
- Increase throughput with cross-Region inference
- Supported Regions and models for inference profiles
- AWS Local Zones locations
- Amazon Bedrock service quotas
- Service tiers for optimizing performance and cost
- Process multiple prompts with batch inference
- Supported Regions and models for batch inference
- Code examples for batch inference
- Amazon Bedrock prompt caching
- Prompt caching for faster model inference
- Inference request parameters and response fields for foundation models
- Boto3 documentation: BedrockRuntime invoke_model
- Document understanding examples in the Amazon Nova User Guide
- Intelligent document processing using Amazon Bedrock and Anthropic Claude
- Solutions for building modern applications with Amazon ElastiCache and Amazon MemoryDB for Redis
- Enable agent memory in the Amazon Bedrock User Guide
- Add memory to your AI agent in the Amazon Bedrock AgentCore Developer Guide
- Converse API examples
Module 5: Customizing Generative AI Responses with RAG
Websites
- Vector database options for RAG use cases
- Vector database comparison for RAG use cases
- Amazon S3 Vectors
- Create a knowledge base in Amazon Bedrock Knowledge Bases
- Set up security configurations for your knowledge base
Module 6: Integrating Open Source Frameworks with Amazon Bedrock
Websites
- DSPy
- LiteLLM: AWS Bedrock
- Haystack: Amazon Bedrock integration
- LangChain introduction
- LlamaIndex: Bedrock Converse
- Mirascope
- Mirascope Lilypad
- LangChain for JavaScript
- What is LangChain?
- LangChain AWS integrations
- LangChain Bedrock integration
- Build an LLM application with chat models and prompt templates
- LangChain ChatPromptTemplate
- LangChain how-to guides
- LangChain integration packages, including langchain-aws
- LangChain DynamoDB integration
- How to add memory to chatbots
- Step 1: Create a table in DynamoDB
Module 7: Evaluating Generative AI Application Components
Websites
- Model evaluation task types in Amazon Bedrock
- Create an automatic model evaluation job in Amazon Bedrock
- Use LLM-as-a-judge model evaluation
- Evaluate the performance of RAG sources using Amazon Bedrock evaluations
Module 8: Implementing Responsible AI
Websites
- Responsible AI on AWS
- Responsible AI in the generative era
- Common prompt injection attacks
- Detect prompt attacks with Amazon Bedrock Guardrails
- Use contextual grounding check to filter hallucinations in responses
- Minimize AI hallucinations and deliver up to 99% verification accuracy with Automated Reasoning checks: Now available
Module 9: Using Tools and Agents in Generative AI Applications
Websites
Research papers
Module 10: Developing Amazon Bedrock Agents
Websites
Amazon Bedrock Flows
- Key definitions for Amazon Bedrock Flows
- Use expressions to define inputs in Amazon Bedrock Flows
- Node types for your flow
- Prompt management in Amazon Bedrock
- Example flows in Amazon Bedrock
- Run Amazon Bedrock Flows code samples
- Amazon Bedrock Knowledge Bases
- What is Amazon Lex V2?
- AI services opt-out policies in AWS Organizations
- FlowNode in the Amazon Bedrock API Reference
Amazon Bedrock Agents
- Amazon Bedrock Agents
- Agents for Amazon Bedrock API Reference
- Agents for Amazon Bedrock Runtime API Reference
- Amazon Bedrock Agents samples on GitHub
- Create an agent with Guardrails for Amazon Bedrock integration (sample)
- Create agents with memory (sample)
Amazon Bedrock AgentCore
- Amazon Bedrock AgentCore
- Introducing Amazon Bedrock AgentCore: Securely deploy and operate AI agents at any scale
- Amazon Bedrock AgentCore documentation
- Amazon Bedrock AgentCore Developer Guide
- Amazon Bedrock AgentCore Control Plane API Reference
- Amazon Bedrock AgentCore Data Plane API Reference
- Host agent or tools with Amazon Bedrock AgentCore Runtime
- Iterate faster with Amazon Bedrock AgentCore Runtime direct code deployment
- Create agent and tool identities with AgentCore Identity
- Amazon Bedrock AgentCore Gateway
- Add memory to your AI agent with AgentCore Memory
- Interact with web applications using Amazon Bedrock AgentCore Browser
- Execute code and analyze data using Amazon Bedrock AgentCore Code Interpreter
- Observe your agent applications on Amazon Bedrock AgentCore Observability
- Configure AgentCore Observability
- View AgentCore observability data
- Amazon Bedrock AgentCore provided observability metrics
- AgentCore Policy
- AgentCore Evaluations
Module 11: Course Wrap-Up
Websites
- AWS Skill Builder
- AWS Classroom training
- AWS Certification
- AWS Certification exam preparation
- AWS Certification exams
- AWS Recertification
- Schedule an Exam
- AWS Ramp-Up Guides
- AWS Workshops
Courses
- Advanced Generative AI Development on AWS
- Building Agentic AI with Amazon Bedrock AgentCore
- Spec-Driven Development with Kiro
- AWS Skill Builder courses
- AWS Builder Labs
Videos
Extra material
Additional diagrams and summaries prepared for this course by deepdives.eu.
- Module 2: Top K and top P inference parameters (PNG)
- Module 7: Evaluation in one diagram (PNG)
- Module 7: RAG evaluation in one diagram (PNG)
- Module 7: RAG evaluation metrics (PDF)
- Module 7: Context metrics explained (PDF)
- Modules 9 and 10: Agent services on AWS in one diagram (PNG)
- Modules 9 and 10: Amazon Bedrock Agents vs. Strands Agents vs. Amazon Bedrock Flows (PNG)
