Key Features

• Build and optimize production-grade RAG pipelines for factual, grounded AI outputs
• Design single-agent and multi-agent architectures for complex enterprise workflows
• Apply evaluation, safety, privacy, and governance frameworks to deploy trustworthy AI at scale

Book Description:

Building Intelligent Systems with LLMs: RAG, AI Agents, and Beyondis a practical guide for engineers, architects, researchers, and technical leaders who want to move from LLM demos to reliable, scalable products. It bridges core AI theory with real implementation practices, showing how modern intelligent systems are designed, evaluated, secured, and operated under real enterprise constraints.

The book covers the full journey from foundational model concepts to advanced system architecture. You will learn how to design and tune Retrieval-Augmented Generation (RAG) pipelines, build autonomous and multi-agent workflows, and implement robust evaluation methods for quality, grounding, hallucination control, and safety. It also provides practical guidance for integrating guardrails, privacy controls, compliance-aware patterns, and operational governance.

With deep technical chapters and hands-on lab-style projects, this book walks through building AI agents, enterprise assistants, production retrieval platforms, vector and hybrid search systems, full DevSecOps delivery pipelines, and LLM safety controls. By the end, you will have the architecture patterns and operational discipline needed to launch trustworthy AI systems in production.

• Understand the evolution from early neural networks to modern large language models
• Design and optimize RAG pipelines for enterprise use cases
• Build single-agent and multi-agent systems for planning and execution
• Evaluate AI outputs for grounding, quality, hallucination risk, and safety
• Implement guardrails, privacy controls, and compliance-ready AI patterns
• Integrate LLM systems with APIs, databases, and external tools
• Operate AI platforms with DevSecOps, observability, release gates, and incident playbooks

This book is for software engineers, AI/ML practitioners, solution architects, researchers, and technical leaders building real-world AI applications. It is ideal for teams moving from prototype to production and for professionals who need reliable, auditable, and scalable LLM systems. Basic familiarity with Python and Generative AI concepts is recommended.

Table of Contents

1. Preface
2. The Evolution of Intelligence: From Perceptrons to Large Language Models
3. How LLMs Work: Transformers, Attention, and Context
4. Training and Fine-Tuning LLMs
5. Prompt Engineering and In-Context Learning
6. Retrieval-Augmented Generation (RAG)
7. What Are AI Agents?
8. Memory, Planning, and Tool Use in Agentic Systems
9. Prompt Engineering in the Age of AI Agents and Vibe Coding
10. Evaluating and Testing AI Systems
11. RAG in Production: Scaling, Performance, and Cost
12. Guardrails, Safety, and Reliability in Production AI
13. Operational Readiness and AI Governance
14. Agent-Oriented Architecture
15. Operational Excellence for AI Systems
16. The Next Wave of RAG
17. Autonomous and Self-Evolving AI Systems
18. AI-Driven Socio-Technical Systems and Global Impact
19. Building AI Agents: Hands-On Project
20. Building RAG Systems: Hands-On Project
21. Building an Enterprise AI Assistant: Hands-On Project
22. Building Multi-Agent Workflows: Hands-On Project
23. RAG + Multi-Agent Integration Project
24. Advanced Vector and Hybrid Retrieval Project
25. Full DevSecOps Pipeline for AI Project
26. LLM Safety Engineering Project
27. Capstone and Deployment Playbook