HANDBOOK OF AI ENGINEERING AND AGENTIC AI : Building Intelligent AI Systems with LLMs, RAG, Multi-Agent Systems, and Production Workflows

Edward Reyland

ISBN 13: 9798191502038

Edité par Independently Published Aug 2026, 2026

A propos de cet article

Neuware.

N° de réf. du vendeur 9798191502038

Détails bibliographiques

• Titre: HANDBOOK OF AI ENGINEERING AND AGENTIC AI : Building Intelligent AI Systems with LLMs, RAG, Multi-Agent Systems, and Production Workflows
• Éditeur: Independently Published Aug 2026
• Date de parution: 2026
• Langue: anglais
• ISBN à 13 chiffres: 9798191502038
• Reliure: Taschenbuch
• État de l'article: Neu
• Dimensions: 279 L x 216 H x 23 P (en millimètres) 279 millimètres de largeur par 216 millimètres de hauteur par 23 millimètres de profondeur
• Poids: 1050 grammes

A propos de ce titre

Synopsis

HANDBOOK OF AI ENGINEERING AND AGENTIC AI
Building an impressive AI prototype is easier than ever. Engineering an AI system that stays reliable, secure, measurable, and economical in production is a much harder problem.
Handbook of AI Engineering and Agentic AI is a rigorous, engineering-first guide to building intelligent systems that move beyond promising demos and perform dependably under real-world conditions. Prompts go brittle. Retrieval misses critical evidence. Structured outputs fail validation. Agents lose state, repeat actions, or exceed intended boundaries. This handbook develops the engineering principles and practical methods needed to address those failures systematically.
Inside, you will learn how to:

• Understand foundation-model behavior, context limits, decoding, caching, and cost trade-offs.
• Engineer prompts, structured outputs, and tool interfaces for dependable applications.
• Build retrieval-augmented systems using chunking, hybrid retrieval, reranking, and advanced retrieval architectures.
• Decide when retrieval, fine-tuning, or both are appropriate—and evaluate results using evidence rather than intuition.
• Design evaluation systems for quality, groundedness, robustness, safety, and cost.
• Build agentic systems that plan, use tools, maintain memory, recover from failures, and incorporate human oversight.
• Design multi-agent architectures in which specialized agents coordinate complex workflows.
• Apply security controls, guardrails, observability, and failure management to production systems.
• Optimize latency and cost while deploying and governing systems at scale.

Agentic AI receives substantial treatment throughout the handbook. Rather than reducing agents to simple prompt loops, the book examines planning, task decomposition, memory, state management, tool use, reflection, permission boundaries, failure recovery, and coordination between specialized agents—the engineering mechanisms required for controlled autonomy.
Each chapter combines learning objectives, first-principles explanations, worked examples, practice problems with worked solutions, and concise summaries, making the book suitable for structured study or professional reference.
Written for AI engineers, machine-learning engineers, software engineers, architects, technical leads, and advanced computing students, the handbook assumes basic code literacy and systems thinking, not a machine-learning research background.
The focus is not on temporary interfaces or short-lived tricks, but on the durable engineering principles behind intelligent systems that can be evaluated, secured, scaled, and operated with confidence.