Autoplay
Autocomplete
Previous Lesson
Complete and Continue
AI Engineering Bootcamp: RAG (Retrieval Augmented Generation) for LLMs
Section 1: Introduction to Retrieval Augmented Generation (RAG) Systems
Course Outline (8:11)
Exercise: Meet Your Classmates and Instructor
Meet Rubber Ducky! Your AI Course Assistant using RAG (6:04)
Link to Your AI Course Assistant
ZTM Plugin + Understanding Your Video Player
[ACTION] Download the Course Resources
Set Your Learning Streak Goal
Section 2: Fundamentals of Retrieval Systems
Overview: Fundamentals of Retrieval Systems (4:04)
Overview of Information Retrieval (5:37)
What is Tokenization? (7:20)
OpenAI Tokenizer (6:12)
Libraries and Data Handling for RAG (6:35)
Tokenization Techniques (6:59)
Preprocessing Steps (9:35)
Types of Retrieval Systems (7:06)
Vector Space Model (TF-IDF) (9:44)
Implementing TF-IDF - Part 1 (7:51)
Implementing TF-IDF - Part 2 (4:15)
TF-IDF Function and Output Analysis (4:58)
Boolean Retrieval Model (6:03)
Preprocessing Steps - Part 2 (6:56)
Setting a Directory (6:04)
Boolean Retrieval Implementation (8:42)
Probabilistic Retrieval Model (7:14)
Probabilistic Retrieval Model Setup (3:32)
Probabilistic Retrieval Model (5:37)
How Google Search Works (11:29)
Key Concepts: Indexing, Querying, and Ranking (7:22)
What Did You Learn in This Section? (2:31)
Let's Have Some Fun (+ More Resources)
Section 3: The Science of Prompt Engineering
ReAct Prompt Engineering (11:52)
Chain of Thought Prompt Engineering (14:24)
Section 4: Generative AI Fundamentals
Overview: Generative AI Fundamentals (1:46)
Introduction to Text Generation (4:09)
Understanding Transformers (12:47)
Rock-Paper-Scissors, Dices and Strawberries (8:24)
Getting a Hugging Face Key (4:19)
Langchain and Hugging Face Setup (13:08)
Basic Text Generation (7:34)
Attention Mechanisms (6:14)
Understanding Generation Model Parameters (6:22)
System Message and Parameters (14:59)
Text Generation with System Message (6:14)
Text Generation with Parameters (10:49)
OpenAI Playground - top P (4:51)
What Did You Learn in This Section? (1:39)
Unlimited Updates
Section 5: The Science of LLMs
LLMs, Few-shot, Scaling and Factuality (14:25)
Section 6: Retrieval-Augmented Generation (RAG) Fundamentals
Overview: RAG Fundamentals (2:38)
Introduction to RAG Architecture (5:17)
Hugging Face Setup (6:29)
Tokenization and Embeddings for RAG (9:02)
FAISS Index: Efficient Similarity Search (4:15)
Building a Retrieval System (4:30)
Developing a Generative Model (7:41)
Implementing the RAG System (7:52)
What Did You Learn in this Section? (3:09)
Implement a New Life System
Section 7: The Science of RAG
LongRAG and LightRAG (16:40)
Section 8: Working with the OpenAI API
Overview: Working with the OpenAI API (3:47)
OpenAI API for Text (8:47)
Setting Up OpenAI API Key (5:49)
OpenAI API Setup (4:31)
Generating Text with OpenAI API (7:02)
OpenAI API Parameters (10:20)
OpenAI API for Images (8:14)
With Image URL (4:54)
Converting Images to Base64 (3:49)
Assess My Python Course Thumbnail (4:48)
What Did You Learn in this Section? (3:50)
Section 9: Project - Customer Acquisition
Project Briefing: Customer Acquisition (6:07)
OpenAI Setup (5:39)
AI Agent System Prompt (8:21)
Processing Images for GenAI (5:24)
Extract Data with GenAI (13:38)
Improving GenAI Extraction (6:18)
GenAI with all Images (6:55)
PDF to Images (10:31)
Wrapping Up the OpenAI GenAI Project (8:16)
Section 10: RAG with OpenAI GPT Models
Overview: RAG with OpenAI GPT Models (4:34)
Case Study Briefing: Cooking Books (4:57)
Converting PDF to Images (9:15)
Reading a Single Image with GPT (12:03)
Enhancing AI with Prompt Engineering (9:10)
Reading All Images in a Dataset (5:07)
Filtering Non-relevant Information (6:03)
Understanding Embeddings in NLP (6:50)
Generating Embeddings (13:56)
Building FAISS Index and Metadata Integration (6:27)
Implementing a Robust Retrieval System (14:41)
Combining Outputs for Enhanced Results (2:56)
Constructing a Generative Model (11:42)
Complete RAG System Implementation (6:41)
How to Improve RAG Systems Effectively? (7:03)
Course Check-In
Section 11: Working With Unstructured Data
Overview: Working With Unstructured Data (3:36)
Introduction to Langchain Library (7:26)
Excel Data: Best Practices for Data Handling (6:41)
Python - Initial Setup for Data Processing (5:47)
Loading Data and Implementing Chunking Strategies (5:13)
Developing a Retrieval System for Unstructured Data (6:10)
Building a Generation System for Dynamic Content (9:12)
Building Retrieval and Generation Functions (9:57)
Working with Word Documents (4:54)
Setting Up Word Documents for RAG (6:17)
Implementing RAG for Word Documents (2:26)
Working with PowerPoint Presentations (4:44)
PowerPoint Setup for RAG (4:11)
RAG Implementation for PowerPoint (3:09)
Working with EPUB Files (4:58)
EPUB Setup for RAG (4:47)
RAG Implementation for EPUB Files (2:22)
Working with PDF Files (4:21)
PDF Setup for RAG (5:51)
RAG Implementation for PDF Files (5:37)
What Did You Learn in This Section? (3:56)
Exercise: Imposter Syndrome (2:55)
Section 12: Multimodal RAG
Overview: Multimodal RAG (3:38)
Introduction to Multimodal RAG (5:58)
Setup and Video Processing (5:23)
Extracting Audio from Video (8:44)
Compressing Audio Files (4:17)
Transcribing Audio with OpenAI Whisper (10:07)
Whisper Model (6:31)
Extracting Frames from Video (5:49)
Introduction to Contrastive Learning (5:14)
Understanding the CLIP Model (5:22)
Tokenizing Text for Multimodal Tasks (8:13)
Chunking and Embedding Text (11:36)
Embedding Images for Multimodal Analysis (8:36)
Understanding Cosine Similarity in Multimodal Contexts (6:46)
Applying Contrastive Learning and Cosine Similarity (10:26)
Visualizing Text and Image Embeddings (11:11)
Query Embedding Techniques (4:12)
Calculating Cosine Similarity for Query and Text (11:47)
GenAI Model Setup for Multimodal Tasks (4:55)
Building a GenAI Model (7:11)
What Did You Learn in This Section? (2:12)
Section 13: Project - Starbucks Financial Data
Project Briefing: Starbucks Financial Data (5:27)
Transcribing Audio with OpenAI Whisper (11:22)
Embedding Transcription with CLIP (7:35)
Converting PDF to Images (5:57)
Embedding Images for Multimodal Analysis (4:58)
Retrieval System (17:13)
Preparing Context (4:59)
Generative System (12:46)
Section 14: RAG with OpenAI File Search
RAG with OpenAI File Search (8:31)
Vector Stores in OpenAI (5:52)
Setting a Vector Store in the OpenAI API (5:46)
Responses Endpoint with File Search (7:27)
RAG with GPT-4.1-mini (6:59)
RAG with System Developper / Messages (5:34)
Section 15: Agentic RAG
Overview: Agentic RAG (2:51)
AI Agents (7:51)
Agentic RAG (5:44)
Setup and Data Loading (9:54)
State Management and Memory in Agentic Systems (7:54)
AgentState Class (4:29)
Greeting the Customer (4:52)
AI Agent that Checks the Question (10:47)
AI Agent that Assesses the Validity of the question (7:22)
Retrieving the Documents (5:46)
Testing the App (7:13)
Generate Answers (9:21)
AI Agent that Improves the Answer (11:13)
Asking User For More Questions (5:29)
Agentic RAG Recap - Key Learnings and Next Steps (6:17)
Section 16: Knowledge Graphs with LightRAG
Game Plan for Knowledge Graphs with LightRAG (2:19)
Knowledge Graphs (7:19)
Knowledge Graphs vs Embeddings (8:49)
LightRAG library update (April 2025)
LightRAG Setup (7:35)
What is LightRAG? (4:40)
Setting the Working Directory (2:28)
Data Prep (4:48)
Naive RAG (6:07)
Implementing LightRAG (6:10)
Knowledge Graph Visualization (8:20)
Local Knowledge Graph Visualization (6:15)
Section 17: RAGAS
Game Plan for RAGAS (1:53)
Assessing RAG with RAGAS (6:13)
RAGAS Setup (3:50)
Embedding and Facebook AI Similarity Search (FAISS) (9:51)
Python - RAG (11:37)
Synthetic Data (3:36)
Generating Synthetic Data (4:59)
Python - Answering Synthetic Dataset (6:28)
ROUGE (Recall-Oriented Understudy for Gisting Evaluation) Score (5:32)
ROUGE (13:49)
LLM-Based Assessment (6:07)
Simple Criteria Score - Part 1 (5:35)
Simple Criteria Score - Part 2 (5:39)
Factual Correctness (5:16)
Rubrics Score (4:52)
Semantic Similarity (4:46)
Factual Correctness (4:57)
Context Precision (3:12)
Semantic Similarity (6:21)
Context Recall (3:11)
Context Precision (5:57)
Response Relevancy (4:36)
Context Recall (4:55)
Response Relevancy (6:22)
Key Learnings and Outcomes: RAGAS (3:17)
Where To Go From Here?
Thank You! (1:17)
Review This Course!
Become An Alumni
Learning Guideline
ZTM Events Every Month
LinkedIn Endorsements
ROUGE (Recall-Oriented Understudy for Gisting Evaluation) Score
This lecture is available exclusively for ZTM Academy members.
If you're already a member,
you'll need to login
.
Join ZTM To Unlock All Lectures