datawhalechina / hello-agents

查看原文 / View Original 编辑 / Edit

Hello-Agents 🤖: Building AI Agents from Scratch

Project Introduction

If 2024 was the “Year of the Hundred Models,” then 2025 has undoubtedly ushered in the “Year of the Agent.” The focus of technology is shifting from training larger foundation models to building smarter agent applications. However, there is a severe lack of systematic, practice-oriented tutorials. To address this, we launched the Hello-Agents project, aiming to provide the community with a guide to building agent systems from scratch, balancing both theory and practice.

如果说 2024 年是”百模大战”的元年，那么 2025 年无疑开启了”Agent 元年”。技术的焦点正从训练更大的基础模型，转向构建更聪明的智能体应用。然而，当前系统性、重实践的教程却极度匮乏。为此，我们发起了 Hello-Agents 项目，希望能为社区提供一本从零开始、理论与实战并重的智能体系统构建指南。

Hello-Agents is a systematic learning tutorial for agents from the Datawhale community. Currently, agent construction is mainly divided into two schools: one is software engineering-oriented agents like Dify, Coze, and n8n, which are essentially process-driven software development where LLMs serve as the data processing backend; the other is AI-native agents, which are truly driven by AI. This tutorial aims to guide you to deeply understand and build the latter—the true AI Native Agent. The tutorial will lead you through the surface of frameworks, starting from the core principles of agents, delving into their core architecture, understanding their classic paradigms, and finally building your own multi-agent applications. We believe that the best way to learn is through hands-on practice. We hope this tutorial will be the starting point for your exploration of the agent world, transforming you from a “user” of large language models into a “builder” of agent systems.

Hello-Agents 是 Datawhale 社区的系统性智能体学习教程。如今 Agent 构建主要分为两派，一派是 Dify，Coze，n8n 这的软件工程类 Agent，其本质是流程驱动的软件开发，LLM 作为数据处理的后端；另一派则是 AI 原生的 Agent，即真正以 AI 驱动的 Agent。本教程旨在带领大家深入理解并构建后者——真正的 AI Native Agent。教程将带领你穿透框架表象，从智能体的核心原理出发，深入其核心架构，理解其经典范式，并最终亲手构建起属于自己的多智能体应用。我们相信，最好的学习方式就是动手实践。希望这本教程能成为你探索智能体世界的起点，能够从一名大语言模型的”使用者”，蜕变为一名智能体系统的”构建者”。

📚 Quick Start

Online Reading: Global Access | Domestic Acceleration - No download required, learn anytime, anywhere. Local Reading: If you wish to read or contribute content locally, please refer to the learning guide below.

在线阅读： 国外访问 | 国内加速 - 无需下载，随时随地学习 本地阅读： 如果您希望在本地阅读或贡献内容，请参考下方的学习指南。

✨ What You Will Gain

  • Datawhale Open Source & Free: Learn all content for free and grow with the community.

  • Understand Core Principles: Deeply understand agent concepts, history, and classic paradigms.

  • Hands-on Implementation: Master the use of popular low-code platforms and agent code frameworks.

  • Self-developed Framework (HelloAgents): Build your own agent framework from scratch based on native OpenAI APIs.

  • Master Advanced Skills: Step-by-step implementation of systematic technologies such as context engineering, Memory, protocols, and evaluation.

  • Model Training: Master Agentic RL, from SFT to GRPO full-process practical training for LLMs.

  • Drive Real-world Cases: Develop comprehensive projects like intelligent travel assistants and cyber towns.

  • Job Interview Preparation: Learn interview questions related to agent positions.

  • Datawhale 开源免费： 完全免费学习本项目所有内容，与社区共同成长

  • 理解核心原理： 深入理解智能体的概念、历史与经典范式

  • 亲手实现： 掌握热门低代码平台和智能体代码框架的使用

  • 自研框架 HelloAgents： 基于 Openai 原生 API 从零构建一个自己的智能体框架

  • 掌握高级技能： 一步步实现上下文工程、Memory、协议、评估等系统性技术

  • 模型训练： 掌握 Agentic RL，从 SFT 到 GRPO 的全流程实战训练 LLM

  • 驱动真实案例： 实战开发智能旅行助手、赛博小镇等综合项目

  • 求职面试： 学习智能体求职相关面试问题

💡 How to Learn

Welcome, future intelligent system builder! Before starting this exciting journey, please allow us to give you some clear guidance. This project covers both theory and practice, aiming to help you systematically master the entire process of designing and developing from single agents to multi-agent systems. Therefore, it is especially suitable for AI developers with some programming foundation, software engineers, students, and self-learners with a strong interest in cutting-edge AI technology. Before learning this project, we hope you have basic Python programming skills and a conceptual understanding of large language models (e.g., knowing how to call an LLM via API). The focus of the project is on application and construction, so you do not need a deep background in algorithms or model training.

欢迎你，未来的智能系统构建者！在开启这段激动人心的旅程之前，请允许我们给你一些清晰的指引。本项目内容兼顾理论与实战，旨在帮助你系统性地掌握从单个智能体到多智能体系统的设计与开发全流程。因此，尤其适合有一定编程基础的 AI 开发者、软件工程师、在校学生 以及对前沿 AI 技术抱有浓厚兴趣的 自学者。在学习本项目之前，我们希望你具备基础的 Python 编程能力，并对大语言模型有基本的概念性了解（例如，知道如何通过 API 调用一个 LLM）。项目的重点是应用与构建，因此你无需具备深厚的算法或模型训练背景。

🤝 How to Contribute

We are an open-source community and welcome any form of contribution!

  • Report Bugs: If you find content or code issues, please submit an Issue.
  • Propose Suggestions: If you have good ideas for the project, feel free to start a discussion.
  • Improve Content: Help improve the tutorial by submitting your Pull Request.
  • Share Practice: Share your study notes and projects in the “Community Selected Contributions.”

我们是一个开放的开源社区，欢迎任何形式的贡献！

  • 报告 Bug： 发现内容或代码问题，请提交 Issue
  • 提出建议： 对项目有好想法，欢迎发起讨论
  • 完善内容： 帮助改进教程，提交你的 Pull Request
  • 分享实践： 在”社区贡献精选”中分享你的学习笔记和项目
← 返回新闻列表 / Back to news list