[Part 01: Getting Started with Langchain]

Chapter 01 Understanding AI Poets and ChatPDF Services
AI poet
ChatPDF
Monetization Method
Technology Stack

Chapter 02 Understanding Langchain and the GPT Model
What is LLM?
What is Langchain?
Why Learn Langchain
Langchain v0.3
Understanding GPT Models with OpenAI Playground

Chapter 03 Setting up the basic development environment
Visual Studio Code Preferences
Python environment settings
Setting environment variables
Langchain environment settings
LLM Chain
Miniconda setup
Langsmith settings

[Part 02: Langchain Basics]

Chapter 04: Building AI to Evaluate Food, Restaurant, and Lodging Reviews
Creating AI to Rating Food Reviews
Building an AI that evaluates restaurant reviews
Building an AI-powered accommodation review evaluation system based on LCEL

Chapter 05: Creating an AI Poet
Service structure to be implemented
LLM chain creation
Streamlit Basics
Front-end implementation
Field deployment

Chapter 06 Creating a Multilingual Email Generator
Ollama Installation and Configuration
Create an email response
Configuring the Streamlit app

[Part 03 Q&A Service Using Document Embedding]

Chapter 07 Creating ChatPDF
Service structure to be implemented
Document Loader
Text Splitter
Embedding
Vector storage
Search engine
Generator
Front-end implementation
Field deployment
Monetizing Web Services
Streaming implementation

Chapter 08: Creating a Bot by Author Hyun Jin-geon
What is the Responses API?
Understanding File Search Tools with OpenAI Playground
Responses API integration
Front-end implementation

[Part 04 Similarity Search Service Using the RAG Technique]

Chapter 09 Creating FAISS Indexes
What is FAISS?
Splitting text data into chunks
Create index

Chapter 10: Implementing Similarity Search with FAISS VectorDB
Query-based similar document search
Retrieving document similarity using embedding vectors

Chapter 11: Implementing RAG-Based Large-Scale Text Search
Create FAISS index
Formatting documents and generating responses

[Part 05: News Search Service Using Advanced RAG Techniques]

Chapter 12: Building a Multiquery-Based News Retrieval System
Multiquery + Unique-union technique
Multiquery + RAG Fusion technique

Chapter 13: Building a Hybrid Search System
Building an Advanced RAG System Using Hybrid Search
Multiquery + Hybrid Search + RAG Fusion Technique
Multiquery + Hybrid Search + RAG Fusion + Streamlit techniques

[Part 06: Integrated Services Using Multimodal Data]

Chapter 14: Building a Multimodal Data RAG System
Multimodal RAG Overview
Multimodal RAG architecture
Installing packages and setting up the JupyterLab environment
Data extraction and segmentation
Multi-vector searcher
Multimodal RAG chain

Chapter 15 FashionRAG: Image-Based Styling Assistant
Understanding the FashionRAG System
Load the Fashionpedia dataset
Base64 encoding

Chapter 16: Creating a Poetry/Novel Generation Service
Configuring Applications with LangServe and FastAPI
Comparison of OpenAI and Ollama models
Building interfaces with Streamlit

[Part 07 Agents Using Langgraph and Agentic RAG]

Chapter 17: Building AI Agents That Use Tools
Main components of the system
Agent and tool integration

Chapter 18: Building AI Agents Using Langgraph
Key features of Langgraph
Create workflow graphs and manage their states
Visualize workflow graphs and interact with agents

Chapter 19: Building an Intelligent Information Retrieval System with Agentic RAG
Agent flow
Defining Agent Status
Create a workflow graph

[Part 08 Collaborative Agents Using CrewAI]

Chapter 20 Multi-Agent Blog Writer
Characteristics of artificial intelligence agents
Agent Definition
Define the task to be performed

Chapter 21: FastAPI, CrewAI-based Blog Content Generator
Agent Definition
Define the task to be performed
Combining CrewAI logic with FastAPI web services

Chapter 22: Building a Blog Service with React Integration
Install Node.js
React project setup
Create the components needed for your project

From Langchain and Langgraph to RAG Fusion and Agentic RAG
Master LLM Development in One Book

This book helps you understand LLM skills naturally through examples rather than complex theories.
The progressively expanding hands-on structure makes it easy for even beginners to develop AI agent services.
In particular, it clearly shows practical examples that can be applied in practice, covering the implementation of multimodal RAG that answers questions by integrating not only text but also tables and images, and the use of the latest VLMs such as GPT-4 Vision and LLaVA.

● Step 1 | First Steps to LLM
Anyone with basic Python knowledge can get started with this guide, from setting up the GPT API to installing Langchain and configuring the development environment.
Step 2 | Create your own AI service that can write poems and read emotions.
Master core concepts with Langchain basics exercises, including AI poets, review rating AI, and multilingual email generators.
Step 3 | Create a 'ChatPDF' that reads and responds to documents
Learn how to monetize your service by building a document Q&A chatbot that analyzes PDFs and answers questions, and even adding API key input and donation buttons.
Step 4 | Building an Advanced Search-Based AI Service with RAG and VectorDB
Embed documents with FAISS and ChromaDB, and implement cutting-edge search technologies such as RAG Fusion and Hybrid Search.
Step 5 | Designing an Agent System with LangGraph and CrewAI
Learn how to handle complex workflows by designing conditional branching, multi-agents, and collaborative agents.
Step 6 | Creating a Complete Service from Web App Deployment, Monitoring, and Monetization
Deploy with Streamlit, track performance with Langsmith, and build a real operational foundation with API key management and sponsorship features.

I recommend this to these people!

● Those interested in creating AI services using LLM
● Those who want to create AI services but find developer-only tools too difficult
● Those who find the official documentation too difficult or have difficulty finding practical examples in RankChain
● Those who want to create their own web service but are unaware of design and deployment
● Those who are interested in not only technology but also service operation and monetization
● Practitioners, planners, marketers, and creators who don't want to fall behind the generative AI trend

A word from beta readers who read it first

● This is the first book you should read if you want to create a service using LLM, and it will serve as a beacon for developers preparing to develop artificial intelligence services.
_Kim Min-gyu
● Clear explanations and practical example code help readers easily build their own AI services. This book is a must-read for anyone seeking to secure a competitive edge in the AI ​​service market.
_Lee Seok-gon
● Each example is realistic and practical enough to be connected to an actual service rather than simply implementing a simple function.
As you follow along, you will naturally grasp the structure of the Langchain and the flow of the AI ​​agent.
_Lee Ji-ah