Designing Enterprise Retrieval, Search, and Intelligent Assistants at Scale
Buch, Englisch, Format (B × H): 155 mm x 235 mm
ISBN: 979-8-8688-3310-6
Verlag: APRESS L.P.
In most organizations, valuable knowledge is scattered across documents—manuals, emails, spreadsheets, and reports—often locked away in formats that are difficult to search, reuse, or scale. This book takes a refreshingly practical view of this challenge, showing how enterprise information can be transformed into structured, reliable systems that power intelligent assistants. Rather than treating AI as a standalone solution, the book positions it as one piece of a larger engineering puzzle.
The book walks the reader through a clear, disciplined approach to building enterprise knowledge platforms. Beginning with why organizations lose knowledge and struggle to find answers, it gradually builds toward a complete system architecture. The early chapters establish the importance of structured documentation, metadata, and retrieval as the foundation for accurate AI responses. From there, the focus shifts to implementation—covering ingestion pipelines, document normalization, chunking strategies, indexing, and hybrid retrieval models that combine traditional search with vector similarity. What makes this book particularly valuable is its emphasis on real-world engineering concerns. Readers learn how to design scalable and secure systems, manage conversational agents with memory and context, and handle governance, permissions, and operational complexity. Along the way, the author shares practical patterns, design principles, and lessons learned from building production systems—including what fails and why.
By the end of the book, readers will have a clear roadmap for turning fragmented enterprise data into a dependable AI knowledge platform. Grounded in decades of hands-on experience, this guide goes beyond theory to offer a realistic, engineering-driven path to building intelligent systems that truly work in practice.
What you will learn:
- Design enterprise AI systems using structured documentation, metadata, retrieval engineering, and hybrid search architectures.
- Convert enterprise documents into AI-ready knowledge platforms capable of powering intelligent assistants.
- Build practical RAG systems using indexing, chunking, vector similarity, and conversational AI workflows.
- Implement governance, security, and memory management to ensure reliable and controlled AI operations.
- Architect scalable, maintainable, and trustworthy enterprise AI platforms grounded in real-world practices.
Who this book is for:
This book is intended for software developers, systems architects, enterprise architects, AI engineers, database professionals, IT managers, knowledge-management professionals, and digital transformation leaders interested in building enterprise AI systems based on structured organizational knowledge.
Zielgruppe
Professional/practitioner
Autoren/Hrsg.
Fachgebiete
Weitere Infos & Material
Part 1 - Why Businesses Struggle to Find Answers.- CHAPTER 1 — Foundations of Corporate Intelligence.- CHAPTER 2 — Why Organizations Struggle to Find Answers.- CHAPTER 3 — Why Generic AI Frequently Fails in Enterprise Environments.- CHAPTER 4 — The Corporate Intelligence System.- Part 2 - Building the Foundation.- CHAPTER 5 — Enterprise System Design Philosophy.- CHAPTER 6 — Enterprise Implementation Patterns.- Part 3: Preparing Enterprise Knowledge.- CHAPTER 7 — Books, Chapters, and Structured Documentation.- CHAPTER 8 — Writing Better Manuals for AI and Search.- CHAPTER 9 — Document Ingestion and Preparation.- CHAPTER 10 — How Enterprise Search Really Works.- Part 4 - Retrieval-Augmented Generation in Practice.- CHAPTER 11 — Metadata Weighting, Synonyms, and Hybrid Retrieval.- CHAPTER 12 — Building the Enterprise AI Chat Agent.- CHAPTER 13 — Security, Governance, and Scalability.- Part 5 - Operating an Enterprise Knowledge Platform.- CHAPTER 14 — Enterprise System Architecture.- CHAPTER 15 — Engineering Lessons and Failed Designs.- APPENDIX 1 — Header and Screen Help Management Code.- APPENDIX 2 — Table-Driven Menu Bar Code.- APPENDIX 3 — Make Your Programs Domain Agnostic Code.
