DAIR.AI

OVERVIEW

This course focuses on building effective and robust retrieval augmented generation (RAG) applications. Students will learn the fundamental building blocks for building RAG systems and the best practices for creating advanced RAG applications. The course also covers advanced concepts such as Agentic RAG systems.

After completing the course, students will have a solid understanding of RAG systems and acquire best practices for building advanced RAG applications in different domains.

PREREQUISITES

If you are not familiar with advanced prompting techniques for LLMs, we strongly recommend completing both the Introduction to Prompt Engineering and Advanced Prompt Engineering courses (also available to all Pro members).

The main tool you will use is Flowise AI (a popular no-code tool to build advanced RAG-based workflows and agentic workflows), therefore, no programming is required. More details and instructions about how to access and install Flowise AI are provided in the course.

SYLLABUS

Course Introduction

• Get a comprehensive overview of Retrieval Augmented Generation (RAG), its fundamental building blocks, and the course objectives.
• Learn how this hands-on course will guide you through building and enhancing complex RAG systems from scratch.

Introduction to RAG

• Gain a foundational understanding of what a RAG system is and how it combines a Large Language Model with external knowledge.
• Learn when to use a RAG system for knowledge-intensive tasks like customer support, and explore common real-world use cases.

Introduction to Flowise AI

• Get started with Flowise AI, the open-source, no-code tool used throughout the course to build, test, and deploy RAG pipelines.
• Follow a step-by-step guide to install Flowise AI locally and prepare your environment for the hands-on projects.

RAG Architecture

• Learn about the core components of a RAG pipeline, including the retriever and the generator.
• Master fundamental concepts like document chunking, embedding models for vector representation, and vector stores that power semantic search.

Build Your First RAG System

• Apply your knowledge to a hands-on project by building your first RAG system from scratch in Flowise AI.
• Create an educational chatbot that can answer questions about a YouTube lecture by loading, chunking, and indexing its transcript in a vector store.

Chat Assistant with RAG

• Build a practical chat assistant for an online knowledge base, using the course's Prompt Engineering Guide as the data source.
• Learn to create a document store by scraping a live website, processing the HTML into Markdown, and indexing it for retrieval.
• Discover how to enable source documents in your RAG system's response to allow users to verify information.

Advanced RAG

• Enhance your RAG systems with advanced techniques like chain-of-thought prompting to improve the factuality and reasoning of your chatbot.
• Learn to use prompt chaining to split the generation process into distinct steps for reasoning, extraction, and refinement.
• Implement query expansion to improve retrieval accuracy by generating multiple variations of a user's question.

Agentic RAG

• Discover Agentic RAG, a cutting-edge approach that uses an agent to route user queries to different tools, including your document store, a calculator, or other chains.
• Learn how function calling enables an agent to interact with external tools and orchestrate complex, multi-step tasks.
• Build an agentic RAG system that upgrades a food chatbot to not only answer questions but also process and confirm user orders.

Deploying RAG Apps

• Learn how to deploy your RAG applications to a live, public URL using the online version of Flowise AI.
• Discover how to configure your deployed chatbot to collect user feedback, allowing for continuous iteration and improvement.

Conclusion

• Recap the key concepts learned, from basic RAG architecture to advanced agentic RAG systems.
• Receive advanced tips on how to continue improving your RAG applications, including model selection, evaluation, and safety considerations.

Here's a subset of companies whose employees have benefitted from our courses:

Accrete, Airbnb, Alston & Bird, Amazon, Apple, Arm, Asana, Bank of America, Belong For Me, Biogen, Brilliant, Carebound, CDM, CentralReach, Centric Software, Chime, Coinbase, Digital Green, DoHQ, Elekta, Fidelity Investments, Fivecast, Fulcrum Labs, Google, Guru, Gretel, Harrison Insights, Intel, Intuit, Jina AI, JPMorgan Chase & Co, Khan Academy, KnowBe4, Lawyer.com, LinkedIn, LionSentry, MagmaLabs, MasterClass, Meta, Metopio, Microsoft, Moneta Health, Oracle, OpenAI, Rechat AI, RingCentral, Salesforce, Scale AI, Scribd, Space-O Technologies, Sun Life, TD Bank, TELUS Corporations, Trilogy, TTEC Digital, UniCredit, VaxCalc Labs, Vendr, Walmart, Wolfram Alpha, Zemoso Technologies, Zeplin

Reach out to [email protected] for any questions and team/student discounts.