AI Engineering Bootcamp: RAG (Retrieval Augmented Generation) for LLMs

This course will teach you how to create more intelligent AI applications using one of the most important techniques in modern artificial intelligence - Retrieval Augmented Generation (RAG). You will learn how to combine Large Language Models (LLMs) with RAG to develop advanced projects such as chatbots, financial analysis systems, and more.

Why is RAG so important?

The limitations of many AI systems are related to their reliance on outdated data from their training samples. RAG addresses this issue by providing access to up-to-date information from external sources, including databases and documents. This makes AI more accurate and useful in real-world scenarios.

Example:

A chatbot in an online store can instantly check the current inventory based on real-time data, instead of relying on static training data, and give you an accurate answer about product availability and delivery times.

What you will learn:

1. Basics of retrieval systems:

• How to prepare textual data for search
• Various search models (Boolean, vector, probabilistic)
• Indexing, queries, and data ranking

1. Basics of generative models:

• Transformer architecture and attention mechanisms
• Data preparation and text model training

1. Introduction to Retrieval-Augmented Generation:

• Combination of search and generation
• Key principles and application of RAG in real tasks

1. Working with OpenAI API:

• API setup and effective use of prompts
• Configuration parameters and their impact on model behavior

1. Implementation of RAG with OpenAI:

• Building fully functional RAG systems
• Integrating search and generation to solve complex tasks

1. Working with unstructured data:

• Processing data from various formats: PDF, Word, PowerPoint, Excel, and images
• Extracting valuable information from texts and multimedia

1. Multimodal RAG systems:

• Using textual and visual data to expand system capabilities
• Integrating different types of data into a single response

1. Agent systems with RAG:

• Building AI agents capable of interacting with users and performing tasks
• Managing agent states and dynamically executing tasks

Why do you need this course?

You will gain practical skills that will allow you to apply RAG in real projects and build scalable AI applications capable of processing complex queries and dynamically providing up-to-date answers.