Developing Generative AI Applications on AWS

This two-day advanced course is designed for software developers seeking to leverage large language models (LLMs) without fine-tuning, using Amazon Bedrock and LangChain. It covers the basics of generative AI, the foundations of prompt engineering, and architecture patterns for building generative AI applications.

Prerequisites

• Completion of AWS Technical Essentials
• Intermediate proficiency in Python

Target Audience

This course is intended for software developers who:

• Want to integrate generative AI models into their applications
• Are interested in Amazon Bedrock and LangChain for generative AI use cases

Delegates will learn how to

• Describe generative AI and its alignment with machine learning
• Identify the business value of generative AI use cases
• Plan and mitigate risks in generative AI projects
• Understand and implement Amazon Bedrock for generative AI applications
• Apply prompt engineering techniques
• Build and secure generative AI applications using Amazon Bedrock and LangChain
• Design and implement architecture patterns for various generative AI use cases

Outline:

Day One

Module 1: Introduction to Generative AI - Art of the Possible

• Overview of machine learning
• Generative AI use cases
• Risks and benefits of generative AI

Module 2: Planning a Generative AI Project

• Steps in planning
• Identifying risks and mitigation strategies

Module 3: Getting Started with Amazon Bedrock

• Introduction to Amazon Bedrock
• Setting up and using Bedrock in the AWS Console
• Hands-on demonstration

Module 4: Foundations of Prompt Engineering

• Basics of prompt engineering
• Advanced techniques and addressing prompt misuse
• Mitigating bias in prompts
• Hands-on demonstration: Prompt fine-tuning and bias mitigation

 

Day Two

Module 5: Amazon Bedrock Application Components

• Overview of application components (e.g., datasets, embeddings)
• Introduction to RAG (Retrieval Augmented Generation)
• Securing applications

Module 6: Amazon Bedrock Foundation Models

• Amazon Bedrock models and methods
• Hands-on lab: Zero-shot text generation

Module 7: LangChain

• Integrating AWS with LangChain
• Using LangChain agents for prompt templates, chat models, and document loaders
• Hands-on lab: Building applications with LangChain

Module 8: Architecture Patterns

• Generative AI architecture patterns
• Hands-on labs: Text summarisation, chatbots, question answering, and code generation using Amazon Bedrock and LangChain

Andre relevante kurs