Enterprise Generative AI on AWS - Couverture souple

Nexley, Devlin

9789011012714: Enterprise Generative AI on AWS

Couverture souple

ISBN 10 : 9011012712 ISBN 13 : 9789011012714

Editeur : Devlin Nexley, 2026

Afficher les exemplaires de cette �dition comportant l'ISBN

0 D'occasion

1 Neuf

De EUR 35,42

Modern generative AI systems are no longer experimental prototypes running behind isolated APIs. They have evolved into large-scale distributed platforms that combine foundation models, retrieval infrastructure, vector search engines, orchestration frameworks, GPU-accelerated inference clusters, and enterprise governance controls. Building these systems reliably requires far more than model access-it demands a deep understanding of architecture, scalability, operational engineering, and infrastructure economics.

This book explores how production-grade generative AI platforms are designed, deployed, and operated on AWS. Rather than focusing on introductory concepts, prompt engineering techniques, or service walkthroughs, it examines the underlying systems that power enterprise AI applications at scale. Readers will learn how large language models interact with retrieval systems, how inference workloads are scheduled across accelerator infrastructure, how vector search platforms support knowledge-intensive applications, and how AI workloads are governed within complex cloud environments.

Beginning with the architectural foundations of modern foundation model systems, the book examines transformer execution mechanics, inference constraints, distributed compute architectures, and the infrastructure layers required to support large-scale AI workloads. It then moves into the design of retrieval-augmented generation systems, embedding pipelines, semantic search infrastructure, and enterprise knowledge architectures that enable models to operate beyond their training boundaries.

The book also explores the emerging domain of agentic systems and autonomous workflows, examining orchestration frameworks, tool-augmented reasoning architectures, multi-agent coordination models, and event-driven AI execution pipelines. These topics are presented through the lens of systems design, emphasizing failure handling, state management, observability, and scalability rather than application-level abstractions.

Operational excellence remains a central theme throughout. Dedicated chapters address reliability engineering, telemetry pipelines, distributed tracing, security architecture, governance frameworks, compliance controls, and platform observability. Particular attention is given to production realities such as prompt injection defense, data protection, workload isolation, model governance, and infrastructure resilience.

Les informations fournies dans la section � Synopsis � peuvent faire r�f�rence � une autre �dition de ce titre.