Using large language models with your companys internal data can be challenging. To create a trustworthy and stable solution, you need a system that is accurate, efficient, and secure. Building a prototype is one thing, but deploying a system that you can scale and maintain at an enterprise level requires a clear strategy. This book shows you how to build production-ready RAG systems that meet business demands.
- Build an enterprise-level RAG system that scales to meet demand.
- Learn to use RAG with SQL databases and internal documentation.
- Create fast and accurate searches for your applications.
- Discover how to prevent AI hallucinations and inaccurate completions.
- Monitor, scale, and maintain RAG systems in a cost-effective way.
Enterprise RAGshows how to build reliable RAG systems for real use in organisations. The book draws on practical experience from real projects. It explains simple ways to improve search, refine questions, and get better results from your system.
After reading this book, you will know how to sidestep common problems and handle challenges like choosing the right LLM. You will be able to build data workflows that maximise accuracy and address cost and performance issues. This book is for software developers who are proficient in Python and want to build reliable RAG solutions.
Free PDF and epub formats plus online reader with AI assistant.
Retrieval Augmented Generation, or RAG, is the gold standard for using domain-specific data, such as internal documentation or company databases, with large language models (LLMs). Creating trustworthy, stable RAG solutions you can deploy, scale, and maintain at the enterprise level means establishing data workflows that maximize accuracy and efficiency, addressing cost and performance problems, and building in appropriate checks for privacy and security. This book shows you how.
It goes beyond the theory and proof-of-concept examples you find in most books and online discussions, digging into the real issues you encounter deploying and scaling RAG in production. In this book, you’ll build a RAG-based information retrieval app that intelligently assesses data from common business sources, chooses the appropriate context for your LLM, and even writes custom SQL queries as needed.
Inside Enterprise RAG you’ll learn:
• Build an enterprise-level RAG system that scales to meet demand
• RAG over SQL databases
• Fast, accurate searches
• Prevent AI “hallucinations”
• Monitor, scale, and maintain RAG systems
• Cost-effective cloud services for AI
About the book
Enterprise RAG teaches you to build production-ready RAG systems. The guide draws from author Tyler Suard's real-world experience developing effective RAG solutions for Fortune 500 companies. Learn to utilize agent-based retrieval, triage logic, query rewriting, and other cutting-edge strategies for effective RAG. Plus, essential tips and advice ensure you can sidestep RAG’s landmines and handle common problems, from picking the right LLM to handling hallucinations and inaccurate completions.
About the reader
For software developers proficient in Python.
About the author
Tyler Suard is a Senior AI Researcher and Developer at a fortune 500 company.
