LLM Development Basics

Day 1: Concept Establishment and Basic Theory

From Traditional AI to Generative Models

🧠

📚 Today's Content

+

•

1.1 LLM Theoretical Basis

Development from traditional AI to generative AI

Basic principles and key technical points of LLM

Overview of mainstream LLM providers and open source ecosystem

Common model categories and typical application scenarios

•

1.2 Core Components and Key Concepts

Basic elements such as Prompt / Model / Agent / Tool / Memory

Agent protocols: MCP, A2A, AG-UI, etc.

Analysis of typical LLM application architecture and case studies

•

1.3 Tools and Experience Demonstration

Demonstration of general tools such as ChatGPT, Claude, Gemini, etc.

Experience typical interaction methods such as MCP, Canvas, Deep Search, etc.

Use GitHub Copilot, Cursor, etc. to experience the application scenarios of LLM in development

•

1.4 Security and Compliance Basics

Overview of potential risks of LLM

Detailed explanation of OWASP LLM Top 10 (2025) and practical suggestions

💡 Learning Objectives

+

✓ Understand the development history of AI from classic methods to generative models
✓ Master the basic working principles and core components of large language models
✓ Understand the characteristics and selection strategies of mainstream LLM providers
✓ Familiar with the typical application scenarios and security risks of LLM

Day 2: Development Framework and Project Preparation

Development Path from Theory to Practice

🔧

📚 Today's Content

+

•

2.1 LLM Project Development Process

How to plan an LLM application project

Complete development process from demand analysis to model integration

Practical path of data preparation, model access, and service deployment

•

2.2 Overview of Mainstream Development Frameworks

Core architecture and comparison of LangChain and LangGraph

Detailed explanation of core modules for Agent construction: Tool, Memory, Executor, etc.

Applicable scenarios for common protocol support: MCP, A2A, AG-UI

•

2.3 Framework Practice

Use the official Demo to get started with the basic usage of LangChain/LangGraph

Realize basic interactive tasks by combining Agent + Tool + Memory

•

2.4 Experimental Project Preparation

Introduce the experimental environment and preset code structure

Clarify experimental goals and task division, and organize development rhythm

💡 Learning Objectives

+

✓ Master the complete development process and planning methods of LLM projects
✓ Understand the characteristics and applicable scenarios of mainstream development frameworks
✓ Learn to build a fully functional Agent system
✓ Familiar with standard protocols and communication mechanisms for LLM applications

Day 3: Comprehensive Practice and Hands-on Experiments

Complete Practice from Concept to Practical Application

⚡️

📚 Today's Content

+

•

3.1 Experiment 1: Basic Chatbot

Objective: Build a minimally viable LLM chatbot

Steps: Prompt writing → API access → message interaction

•

3.2 Experiment 2: Simple Agent Construction

Objective: Build an Agent with basic reasoning capabilities

Steps: Define Agent → Integrate tools → Context management

•

3.3 Experiment 3: External Tool Invocation

Objective: Implement Agent to call external tools (such as search, calculator)

Steps: Define Tool → Build invocation link → Integrate invocation results

•

3.4 Experiment 4: MCP Server Practice

Objective: Build an Agent service that follows the MCP protocol

Steps: Protocol understanding → Service construction → Interface joint debugging and testing

•

3.5 Summary and Q&A

Review the key knowledge points of the three days

Typical question answering and suggestions for subsequent learning paths

💡 Learning Objectives

+

✓ Master the complete development process of LLM applications through actual projects
✓ Learn to build complex multi-Agent collaboration systems
✓ Master various tool integration and invocation skills
✓ Understand the application of the MCP protocol in actual projects
✓ Establish a complete LLM project development framework and best practices

Syllabus

Day 1: Concept Establishment and Basic Theory

Day 2: Development Framework and Project Preparation

Day 3: Comprehensive Practice and Hands-on Experiments

📦 Appendix

Day 1: Concept Establishment and Basic Theory

📚 Today's Content

1.1 LLM Theoretical Basis

Development from traditional AI to generative AI

Basic principles and key technical points of LLM

Overview of mainstream LLM providers and open source ecosystem

Common model categories and typical application scenarios

1.2 Core Components and Key Concepts

Basic elements such as Prompt / Model / Agent / Tool / Memory

Agent protocols: MCP, A2A, AG-UI, etc.

Analysis of typical LLM application architecture and case studies

1.3 Tools and Experience Demonstration

Demonstration of general tools such as ChatGPT, Claude, Gemini, etc.

Experience typical interaction methods such as MCP, Canvas, Deep Search, etc.

Use GitHub Copilot, Cursor, etc. to experience the application scenarios of LLM in development

1.4 Security and Compliance Basics

Overview of potential risks of LLM

Detailed explanation of OWASP LLM Top 10 (2025) and practical suggestions

💡 Learning Objectives

Day 2: Development Framework and Project Preparation

📚 Today's Content

2.1 LLM Project Development Process

How to plan an LLM application project

Complete development process from demand analysis to model integration

Practical path of data preparation, model access, and service deployment

2.2 Overview of Mainstream Development Frameworks

Core architecture and comparison of LangChain and LangGraph

Detailed explanation of core modules for Agent construction: Tool, Memory, Executor, etc.

Applicable scenarios for common protocol support: MCP, A2A, AG-UI

2.3 Framework Practice

Use the official Demo to get started with the basic usage of LangChain/LangGraph

Realize basic interactive tasks by combining Agent + Tool + Memory

2.4 Experimental Project Preparation

Introduce the experimental environment and preset code structure

Clarify experimental goals and task division, and organize development rhythm

💡 Learning Objectives

Day 3: Comprehensive Practice and Hands-on Experiments

📚 Today's Content

3.1 Experiment 1: Basic Chatbot

3.2 Experiment 2: Simple Agent Construction

3.3 Experiment 3: External Tool Invocation

3.4 Experiment 4: MCP Server Practice

3.5 Summary and Q&A

💡 Learning Objectives

📦 Appendix