RAG-Driven Generative AI - Second Edition: Build MAS-RAG with DualRAG, GraphRAG, multimodal video pipelines, and Oracle Database 23ai

Building MAS-RAG (multi-agent AI systems for RAG) that reason over real-world data using hybrid retrieval and scalable architectures for production use.

Free with your book: DRM-free PDF version + access to Packt's next-gen Reader*

Key Features

• Master DualRAG by combining vector search with SQL filtering over structured enterprise data
• Implement GraphRAG, Spatial-RAG, and vector search natively in Oracle Database 23ai
• Build multimodal video pipelines with human-feedback loops and fine-tuned models

Book Description

Stop moving your data to the AI. This second edition defines a revolutionary architectural shift: bringing the AI to the data. By using Oracle Database 23ai as a converged engine in this book, you will architect Sovereign AI systems that eliminate the fragmentation, latency, and massive security risks inherent in traditional data extraction.

You’ll work with DualRAG, synchronizing unstructured vector semantics with the deterministic truth of structured SQL, Graph, and Spatial retrieval. This allows your systems to reason over verified corporate data rather than probabilistic guesses, reducing hallucinations at the source. Moving beyond simple pipelines, you’ll also build MAS-RAG (multi-agent systems for RAG), where autonomous agents coordinate across hybrid retrieval workflows, multimodal video pipelines, and graph-based knowledge structures.

Designed for developers and architects, these blueprints transform disconnected data silos into a unified engine to architect autonomous enterprise intelligence that scales with RLHF and model fine-tuning. By the end of the book, you’ll be able to design and deploy enterprise AI systems that combine retrieval, reasoning, and structured data to build reliable generative AI applications.

*Email sign-up and proof of purchase required

What you will learn

• Bring intelligence directly to the data within Oracle Database 23ai
• Defeat hallucinations and data poisoning with DualRAG, synchronizing vector semantics with structured SQL
• Build MAS-RAG pipelines with Planner, Agent Registry, and MCP-standardized sovereign agents
• Engineer an inference-time router using hybrid adaptive RAG to switch between reasoning, retrieval, and human feedback
• Fuse vector similarity, Oracle Spatial, and SQL Property Graph traversal into a converged hyper-query
• Multimodal video RAG with version-controlled schema registry and semantic vector search over visual assets

Who this book is for

This book is for AI engineers, ML engineers, data scientists, and MLOps professionals who want to build production-ready generative AI systems grounded in enterprise data. It will also benefit solutions architects, database engineers, and software developers looking to integrate large language models with structured and unstructured data sources using modern retrieval architectures. Readers should be comfortable with Python and have a basic understanding of machine learning concepts. Prior experience with generative AI or vector databases will help you get the most out of this book.

Table of Contents

1. Why Retrieval-Augmented Generation?
2. RAG Embeddings in Oracle Vector Stores
3. Building a Live Recruiter Agent
4. Building Sovereign Enterprise Agents
5. Building a Universal Context Engine
6. Operationalizing the Universal Context Engine
7. Empowering AI Models by Fine-Tuning RAG Data
8. Boosting RAG Performance with Human Feedback
9. Building a Conversational RAG Agent
10. Building an Agent with Spatial-RAG and GraphRAG
11. Scaling AI Workloads with Oracle Exadata
12. The Autonomous Database Architect

Denis Rothman graduated from Sorbonne University and Paris-Diderot University, designing one of the very first word2matrix patented embedding and patented AI conversational agents. He began his career authoring one of the first AI cognitive Natural Language Processing (NLP) chatbots applied as an automated language teacher for Moet et Chandon and other companies. He authored an AI resource optimizer for IBM and apparel producers. He then authored an Advanced Planning and Scheduling (APS) solution used worldwide.