In 2024, we were discussing how to write better Prompts. In 2025, the industry's focus has completely shifted to Agents.
Among the myriad of Agent frameworks and platforms, Hello-Agents, initiated by the Datawhale community, stands out as a unique project. It is not a plug-and-play SaaS platform, but a systematic educational engineering project designed to teach you how to understand, design, and implement true "AI Native" agents from the bottom up.
What is Hello-Agents?
Hello-Agents is an open-source educational project led by the Datawhale community. Its core objective is to break open the "black box" of Agent development, allowing developers to master the core architecture of agents starting from foundational logic.
Unlike platforms that focus on workflow orchestration (like Dify or Coze), Hello-Agents emphasizes autonomous reasoning, dynamic planning, and self-evolution. It not only covers basic theory but also features a lightweight experimental framework, HelloAgents, for learners to validate their ideas through code.
Key Highlights
- Systematic Learning Path: From foundational theory to engineering implementation and advanced extensions (like RLHF, MCP), the project provides a complete progression curve.
- Theory Meets Practice: Every chapter is accompanied by detailed core code implementation. You are not just learning where to click buttons, but how to hand-write a ReAct loop or a memory system.
- Embracing Modern Standards: It is one of the first comprehensive tutorials in China to deeply integrate and explain the Model Context Protocol (MCP), keeping pace with international technical frontiers.
- Deep Dive into Agentic RL: Covers the evolution from SFT to GRPO, exploring how to empower Agents with stronger reasoning capabilities through reinforcement learning.
Technical Breakdown: How to Build an Agent?
Hello-Agents decomposes agent construction into several core elements:
1. Reasoning and Planning (Thinking Loop)
It provides in-depth explanations of mainstream paradigms like ReAct (Reason + Act), Plan-and-Solve, and Reflection. It teaches you how to move beyond LLMs just outputting text to having them "observe" the environment and "think" about the next step.
2. Tool Calling and MCP
It's more than just simple Function Calling. The project highlights the Model Context Protocol (MCP) proposed by Anthropic—a standardized open protocol that allows Agents to connect to various data sources and local tools in a unified way.
3. Memory System
Explores how to build sensory-capable memory systems, including long-term memory enhanced by RAG and session management based on state persistence.
4. RL and Self-Evolution
This is the most advanced part of the project. It discusses the principles of reasoning models represented by OpenAI o1 and how Agentic RL technology allows agents to self-correct and optimize through repeated practice.
Who is Hello-Agents for?
- Developers: Engineers who are already familiar with Python and LLM basics but want to master Agent system architecture and low-level implementations.
- Students/Researchers: Learners who want to systematically study the latest research findings and engineering techniques in the field of agents.
- Product Managers: Those who want a deep understanding of AI Native product logic to design more competitive Agent applications.
Conclusion
In 2025, Agents have evolved from "software engineering logic" to "AI Native logic." Hello-Agents is the perfect ticket to this new world. It doesn't just teach you how to write code; it teaches you a new way of thinking—how to build digital lives that can reason, learn, and act autonomously.
- GitHub: https://github.com/datawhalechina/hello-agents
- Online Tutorial: https://datawhalechina.github.io/hello-agents/