A systematic learning tutorial on AI agents from the Datawhale community, covering foundational theory to practical application. It enables developers to transition from language model users to intelligent system builders.
One-Minute Overview#
"Hello-Agents" is a comprehensive open-source tutorial covering both theory and practice of AI intelligent systems, designed for AI developers and engineers. It starts from basic concepts and progressively delves into multi-agent system development, enabling you to not only proficiently use existing tools but also build your own agent frameworks from scratch.
Core Value: Cultivates full-stack capabilities in designing and developing intelligent systems through a combination of theoretical knowledge and practical implementation.
Getting Started#
Installation Difficulty: Low - This is a tutorial resource that requires no installation, supports online reading and local code execution
# Clone the project code
git clone https://github.com/datawhalechina/hello-agents.git
Is this suitable for me?
- ✅ AI Developers: Want to transition from using language models to building agent systems
- ✅ Software Engineers: Looking to integrate LLMs into practical applications
- ✅ Students: Systematically learn AI agent technology and practices
- ❌ Those with no programming experience: Need to master Python basics first
- ❌ Users seeking ready-to-use tools: This is a learning project, not a ready-made tool
Core Capabilities#
1. Agent Fundamentals and Theory - Building Knowledge Foundation#
- In-depth understanding of agent definitions, types, paradigms, and applications Practical Value: Establishes a systematic knowledge framework for agents, laying theoretical groundwork for subsequent practice
2. Implementing Classic Agent Paradigms - Mastering Core Methods#
- Step-by-step implementation of ReAct, Plan-and-Solve, Reflection and other classic paradigms Practical Value: Deep understanding of agent working principles, laying foundation for developing complex agents
3. Building Your HelloAgents Framework - Capability to Build from Scratch#
- Construct your own agent framework from zero based on OpenAI's native API Practical Value: Master agent framework design principles, possessing the ability to "build wheels"
4. Multi-Agent System Design - Building Complex Applications#
- Learn advanced technologies like memory and retrieval, context engineering, and agent communication protocols Practical Value: Ability to design and implement complex multi-agent collaborative systems
5. Real Project Practice - Solving Actual Problems#
- Develop projects like intelligent travel assistants, automated research agents, and Cyber Town Practical Value: Apply theoretical knowledge to real scenarios, building agent systems that solve practical problems
Tech Stack and Integration#
Development Language: Python Main Dependencies: OpenAI API, LangGraph, AutoGen, AgentScope, Transformer Integration Method: Combines tutorials with practical code, providing complete example projects
Ecosystem and Extensions#
- Community Contributions: Welcome contributions of practice cases, study notes, and interview questions
- Extended Content: Provides "Extra-Chapter" to feature selected community content
- Continuous Updates: Planning bilingual video courses for more detailed practical guidance
Maintenance Status#
- Development Activity: Active project with regular content updates and high community participation
- Recent Updates: Content continuously iterated, with new chapters and cases added regularly
- Community Response: Supported by Datawhale community with完善的 issue and PR mechanisms
Commercial and Licensing#
License: CC BY-NC-SA 4.0
- ✅ Commercial Use: Not allowed (non-commercial use only)
- ✅ Modification: Allowed (share alike)
- ⚠️ Restrictions: For non-commercial use only, must attribute the original author
Documentation and Learning Resources#
- Documentation Quality: Comprehensive, including complete tutorial content, code examples, and project cases
- Official Documentation: https://github.com/datawhalechina/hello-agents
- Example Code: Provides complete project code that supports local execution and modification