No. 360 of 500 No. 40 in Generative AI & LLMs

Generative AI on AWS

Building Context-Aware Multimodal Reasoning Applications

By Chris Fregly, Antje Barth and Shelbee Eigenbrode

Print length
309 pages
Publisher
O'Reilly Media
Reading level
Intermediate
ISBN-13
978-1098159221
Published
December 19, 2023
Written for
ML engineers, developers and technical leads building generative AI applications on AWS.

Synopsis

Written by three AWS generative AI specialists who also taught the DeepLearning.AI and AWS course Generative AI with Large Language Models, this O'Reilly book walks through the life cycle of a generative AI project, from choosing a use case to running a model in production. Early chapters cover the transformer-based foundation models behind tools such as ChatGPT, prompt engineering and in-context learning, then the compute realities of large models, including quantization and distributed computing. The middle of the book focuses on adapting models: full fine-tuning and evaluation, parameter-efficient methods such as LoRA, and reinforcement learning from human feedback to align outputs with human preferences.

Later chapters address deployment optimization, retrieval-augmented generation and agents for grounding responses in external data, multimodal models that combine text and images, controlled image generation and fine-tuning with Stable Diffusion, and Amazon Bedrock as a managed service. Examples run on Amazon SageMaker and Bedrock alongside open-source libraries, and a companion GitHub repository organizes code by chapter. Because the book appeared in late 2023, its model choices and service features predate later frontier models and agent frameworks, but its explanations of fine-tuning, alignment and retrieval remain sound. Readers gain a working map of the techniques involved in building generative applications and a practical sense of how they fit together on one cloud platform.

Scorecard

Editorial score

How it scored

Editorial score: 68.6 / 1005 weighted criteria

  • Influence30%

    19.5/30
  • Authority25%

    18.8/25
  • Practical value20%

    15.0/20
  • Currency15%

    8.3/15
  • Readability10%

    7.0/10
  • Editorial score

    68.6/100

How the rank was computed

Rank No. 360 of 500 comes from a composite of 51.1: observable signals 20.2 of 55 points plus editorial review 30.9 of 45 points.

All 11 inputs
InputPoints
Observable signals20.2/55
Reach (Amazon ratings volume)7.3/16.5
Consensus across best-of lists5.5/13.8
Awards and bestseller evidence0.0/8.3
Reader rating, sample-adjusted5.3/8.3
Amazon sales rank0.4/5.5
Recency1.7/2.8
Editorial review30.9/45
Author authority8.4/11.3
Influence8.8/13.5
Practical value6.8/9.0
Readability3.2/4.5
Currency3.7/6.8

Each criterion is scored from 1 to 10 and weighted, and the points add up to the editorial score. The rank combines observable signals (55 percent) with this score (45 percent). Read the methodology.

Verdict

A solid practitioner's tour of the 2023 foundation-model lifecycle on AWS, valuable for fundamentals though its service details age quickly.

Strengths

  • Covers the full model lifecycle end to end
  • Authors taught a widely used LLM course
  • Companion code organized by chapter

Keep in mind

Heavy reliance on AWS services limits portability, and 2023-era models, APIs and Bedrock features have changed substantially since publication.

Key takeaways

  1. Most generative AI projects should begin with prompt engineering and in-context learning before investing in fine-tuning, which adds cost and operational complexity.
  2. Parameter-efficient fine-tuning methods such as LoRA adapt large models to specific tasks at a fraction of the compute and storage of full fine-tuning.
  3. Reinforcement learning from human feedback aligns model outputs with human preferences such as helpfulness and harmlessness, using a reward model trained on human rankings.
  4. Retrieval-augmented generation grounds model responses in current, domain-specific data, reducing hallucinations without retraining the underlying model.

Notable facts

  • The three authors, with Mike Chambers, developed and delivered the DeepLearning.AI and AWS course Generative AI with Large Language Models, launched in June 2023 under Andrew Ng's guidance. Source (opens in a new tab)

About the author

Chris Fregly, Antje Barth and Shelbee Eigenbrode

When the book appeared, Chris Fregly and Shelbee Eigenbrode were Principal Solutions Architects for generative AI at Amazon Web Services, and Antje Barth was AWS's Principal Developer Advocate for generative AI. All three were instructors on the DeepLearning.AI and AWS Coursera course Generative AI with Large Language Models, and Fregly and Barth previously coauthored O'Reilly's Data Science on AWS.

Questions

What does Generative AI on AWS cover?
The book follows the generative AI project life cycle: selecting use cases and models, prompt engineering, quantization and distributed computing, full and parameter-efficient fine-tuning, reinforcement learning from human feedback, deployment optimization, retrieval-augmented generation and agents, multimodal and Stable Diffusion models, and Amazon Bedrock. Examples run on AWS services, with companion code on GitHub organized into twelve chapters.
Is Generative AI on AWS still current in 2026?
Partly. Its explanations of fine-tuning, LoRA, RLHF and retrieval-augmented generation remain useful foundations. The specific models, SageMaker workflows and Bedrock features describe the late-2023 landscape, and newer frontier models, agent tooling and managed services have since changed the details. Use it for concepts and lifecycle thinking, and check current AWS documentation for implementation specifics.
Do I need to use AWS to benefit from Generative AI on AWS?
Not entirely. Much of the material, including prompt engineering, fine-tuning methods, RLHF and retrieval-augmented generation, explains techniques that apply on any platform. The hands-on examples, deployment guidance and the Amazon Bedrock chapter assume AWS, so teams on other clouds will need to translate those sections into equivalent tools and services.

Sources

As an Amazon Associate, Best AI Books earns from qualifying purchases. Best AI Books is sponsored by Iternal.ai, whose founder and CEO, John Byron Hanby IV, wrote The AI Strategy Blueprint, our #1 ranked book. How we rank books.