Amazon Bedrock Customization, Optimization & Automation

Overview

Grow generative AI expertise with this course focusing on customizing, optimizing, and automating AI solutions using Amazon Bedrock. This course is designed for developers who want to fine-tune their AI applications for peak performance and efficiency. You'll begin by exploring model customization techniques, including fine-tuning and continued pre-training. Learn how to adapt foundation models to your specific use cases, enhancing their performance on domain-specific tasks. The course then dives into advanced optimization strategies. You'll work with Bedrock Evaluation Jobs to assess and compare model performance, implement prompt caching for improved response times, and utilize prompt routing for efficient model selection. In the automation section, you'll discover how to streamline AI workflows using Bedrock Data Automation. This tool will enable you to process and transform large datasets. Throughout the course, you'll work in hands-on labs and real-world scenarios, applying these advanced techniques to solve complex AI challenges. By the conclusion of the course, you'll be designing, implementing, and maintaining AI solutions, stretching the limits of what's possible with generative AI on AWS. Please note: The hands-on exercises are optional and require access to your own AWS account. Completing these activities may result in minimal usage charges.

Syllabus

Module 1: Improving Foundation Model Results

This module explores how developers can improve the output of foundation models using customization techniques in Amazon Bedrock. You will learn about fine-tuning with labeled data, continued pretraining with domain-specific content, and model distillation for cost-effective performance. The module also introduces LangChain to enhance AI workflows. You will gain practical knowledge to tailor models to their specific use cases and boost relevance and accuracy.

Module 2: Using Foundation Models for Efficiency

In this module, you will learn how to improve the efficiency and scalability of generative AI applications using Amazon Bedrock. You will use Bedrock Evaluation Jobs for comparing model responses, and apply prompt caching and routing to optimize performance. The module also covers automation techniques using Bedrock Data Automation and Amazon Q Developer on the command line. By the end, you will be equipped to streamline inference, automate tasks, and make intelligent deployment decisions.