Part 1 │ Understanding LLM Trends

Chapter 1: The Emergence of LLM and the Transformation of the AI ​​Ecosystem
__1.1 The Birth of LLM
__1.2 The Evolution of LLM: Multimodal LLM
__1.3 Development of LLM: Scaling up the model size
__1.4 The Evolution of LLM: Expanding the Role of Open Source and the Community

Chapter 2: The Evolution of LLM and the Emergence of AI Agents
__2.1 The emergence of o1
____2.1.1 Comparison of GPT-4o and o1
____2.1.2 Understanding o1 through scenarios
__2.2 The emergence of AI agents

Part 2 │ Understanding AI Agents

Chapter 3: The Concept of AI Agents
__3.1 What is an AI agent?
__3.2 Comparison of LLM, RAG, and AI agents

Chapter 4: AI Agent Configuration and Operation
__4.1 AI Agent Components
__4.2 How AI Agents Work
__4.3 Agent Types

Chapter 5 AI Agent Design Patterns
__5.1 Response Patterns
__5.2 Planning Patterns
__5.3 Tool Usage Patterns
__5.4 Multi-Agent Pattern
__5.5 CoT prompting pattern

Chapter 6 AI Agent Framework
__6.1 Autogen
__6.2 Langchain
__6.3 Langgraph
__6.4 Crew AI
__6.5 Rama Index
__6.6 AutoGPT

Part 3: Using AI Agents

Chapter 7: Preparing to Use AI Agents
__7.1 Configuring the Colab Environment
____7.1.1 Connecting to Colab
____7.1.2 Using Colab
____7.1.3 Accessing Google Drive
__7.2 API Key Issuance
____7.2.1 Generating an OpenAI API Key
____7.2.2 Generating a Tavily API Key
____7.2.3 Generating a Langsmith API Key
____7.2.4 Generating a Serper API Key

Chapter 8: Implementing AI Agents
__8.1 Langchain Agent
____8.1.1 How the Langchain AI Agent Works
____8.1.2 Implementing an Agent with Langchain
__8.2 AutoGPT
__8.3 Autogen
____8.3.1 Understanding the Autogen Concept
____8.3.2 Implementing an Agent with Autogen
__8.4 Rama Index
____8.4.1 Understanding the Rama Index Concept
____8.4.2 Implementing an Agent with RamaIndex
__8.5 Crew AI
____8.5.1 Understanding the Crew AI Concept
____8.5.2 Creating an Agent with Crew AI
__8.6 Using Langgraph
____8.6.1 Understanding the Langgraph Concept
____8.6.2 Creating an Agent with LangGraph

Chapter 9: Agent Debugging and Evaluation Using Langsmith
__9.1 What is Langsmith?
__9.2 Using Langsmith
____9.2.1 Debugging
____9.2.2 Integrating LangGraph with Debugging
____9.2.3 Evaluating Performance
____9.2.4 Monitoring

Part 4 │ M365 Co-Pilot Agent

Chapter 10: M365 Copilot and M365 Copilot Agent
__10.1 M365 Copilot
____10.1.1 M365 Copilot Features
____10.1.2 How Copilot Works
__10.2 M365 Copilot Agent

Chapter 11 Comparing M365 Copilot Agents and AI Agents
__11.1 Limitations of M365 Copilot Agent
__11.2 M365 Copilot Agent vs. AI Agent

Search

Anyone can quickly and easily learn AI agents and utilize various frameworks!
An introduction to AI agents: learn through hands-on implementation! Quickly cover concepts, applications, and practice in one book!

What this book covers: AI agent concepts & related frameworks
Anyone can learn AI agents and create them using various frameworks.
With this one book, you can learn more efficiently by understanding the core concepts of AI agents, mastering frameworks for utilizing these concepts, and implementing various AI agents using them.

[concept]

ㆍ Explore LLM trends and learn about the concepts and design patterns of AI agents.
ㆍ Let's gain a deeper understanding of the changes in the LLM and AI ecosystems and consider scalability.
ㆍ We will examine how to implement LLM and AI agents and what services can be developed.

[Framework]

ㆍ Introducing various frameworks such as LangChain, LangGraph, RamaIndex, and LangSmith.
ㆍ Implement AI agents that can actually be used in each framework.
ㆍ Compare AI agents and Microsoft 365 (M365) Copilot agents.

This book's various practical scenarios: Building AI agents using eight frameworks.
Learn how to implement various AI agents in frameworks such as LangChain, LangGraph, LangSmith, AutoGen, AutoGPT, CrewAI, RamaIndex, and M365 Copilot.
It consists of basic examples that work in Koreb so that anyone can easily follow along.

ㆍ [Langchain] Zero-shot ReAct, Conversational ReAct, Self-ask with search, ReAct docstore
ㆍ [Langgraph] Information retrieval using Tavily, ReAct agent, RAG & search agent, multi-agent
ㆍ [Crew AI] Data search and content creation, searching for data on specific YouTube channels
ㆍ [AutoGen] Q&A Agent, Creating Two or More Assistant Agents
ㆍ [AutoGPT] Using the AutoGPT agent as a 'code mentor'
ㆍ [RamaIndex] Creating a PDF document search and math calculation agent
ㆍ [Langsmith] Debugging, Performance Evaluation, and Monitoring
ㆍ [M365 Copilot] Comparative Analysis of M365 Copilot and AI Agents

[Author's Preface]

Following LLM, interest in AI agents is heating up.
Why are people turning to AI agents? Automation is key. LLM itself, as a language model, specializes in content generation.
Of course, human activity begins with content creation.
Create content to create promotional videos, send emails, or give lectures.
Alternatively, use RAG to check the delivery status of your order or view available flights.
But is it possible to execute tasks like booking airline tickets using just an LLM? This can be done using an AI agent.
In this book, we will cover the most basic methods for implementing such AI agents.
And we introduce various frameworks that can implement AI agents.
We introduce how to implement AI agents in various frameworks, including LangChain, LangGraph, AutoGPT, and AutoGen.