An introduction to retrieval-augmented generation

Retrieval-augmented generation is rapidly becoming the leading Generative AI architecture. While prompting LLMs is powerful, LLMs don't know a lot about your business or its challenges. RAG solves that problem. With RAG, you use the massive powers of LLMs like OpenAI's GPT4 with your corporate information.

The technology powers many applications, from support chatbots to knowledge search, and it should become part of your toolset. Join Domino's John Alexander as he guides you toward your first steps in harnessing this powerful technology for your projects.

What you will learn

1. Introduction: An overview of Domino Data Lab and an introduction to RAG.
2. Theoretical Foundations: We'll review the basics of Natural Language Processing (NLP) and cover various retrieval techniques. We will also explain generative model mechanics to ensure a solid understanding of how RAG fits.
3. How RAG Works: We will explore RAG in detail to better understand how retrieval and generative components interact. We will present real-world case studies and examples to show RAG's practical applications. RAG-based development can be harsh without help. We will, therefore, review the tools and technologies used in building RAG systems.
4. Applications and Challenges: We will finally shift our focus to practical applications and look at RAG-powered chatbots and search engines. Just as importantly, we will review RAG's limits.