Ali Qianwen Launches China's First Full-Cycle College Entrance Exam Volunteer Filling Agent

If 2024 was the first year of the "Hundred Models War", then 2025 undoubtedly marks the beginning of the "Agent Year". The focus of technology is shifting from training larger foundation models to building smarter agent applications. However, systematic and practice-oriented tutorials are currently extremely scarce. To this end, we launched the Hello-Agents project, hoping to provide the community with a guide to building agent systems from scratch, emphasizing both theory and practice.

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

If you wish to read locally or contribute content, please refer to the learning guide below.

- 📖 Datawhale Open Source Free: Learn all content of this project for free and grow together with the community

- 🔍 Understand Core Principles: Deeply understand the concepts, history, and classic paradigms of agents

- 🏗️ 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 OpenAI native API

- ⚙️ 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 of LLMs

- 🚀 Drive Real Cases: Practical development of comprehensive projects such as intelligent travel assistants and cyber towns

- 📖 Job Hunting Interview: Learn interview questions related to agent job hunting

Everyone is welcome to contribute unique insights and practical summaries gained from learning Hello-Agents or Agent-related technologies to the community selection in the form of PRs. If the content is independent of the main text, it can also be submitted to Extra-Chapter! We look forward to your first contribution!

This Hello-Agents PDF tutorial is completely open source and free. To prevent various marketing accounts from adding watermarks and selling it to beginners of multi-agent systems, we have specially added a Datawhale open source logo watermark to the PDF file in advance that does not affect reading. Please understand.

Hello-Agents PDF : https://github.com/datawhalechina/hello-agents/releases/latest/

Hello-Agents PDF Domestic Download Address : https://www.datawhale.cn/learn/summary/239

Welcome, future intelligent system builder! Before starting this exciting journey, please allow us to give you some clear guidance.

The content of this project balances theory and practice, aiming to help you systematically master the entire process of design and development from single agents to multi-agent systems. Therefore, it is especially suitable for AI developers, software engineers, and college students with some programming foundation, as well as 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 basic conceptual understanding of large language models (for example, knowing how to call an LLM via API). The focus of the project is application and construction, so you do not need a deep background in algorithms or model training.

The project is divided into five major parts, each of which is a solid step towards the next stage:

- Part One: Agent and Language Model Basics (Chapter 1 ~ Chapter 3): We will start with the definition, types, and development history of agents, sorting out the context of the concept of "agent". Subsequently, we will quickly consolidate the core knowledge of large language models, laying a solid theoretical foundation for your practical journey.

- Part Two: Building Your Large Language Model Agent (Chapter 4 ~ Chapter 7): This is the starting point for your hands-on practice. You will implement classic paradigms such as ReAct with your own hands, experience the convenience of low-code platforms such as Coze, and master the application of mainstream frameworks such as Langgraph. Finally, we will also guide you to build your own agent framework from scratch, giving you both the ability to "use wheels" and "build wheels".

- Part Three: Advanced Knowledge Expansion (Chapter 8 ~ Chapter 12): In this part, your agent will "learn" to think and collaborate. We will use the self-developed framework from Part Two to deeply explore core technologies such as memory and retrieval, context engineering, and agent training, and learn communication protocols between multi-agents. Finally, you will master professional methods for evaluating agent system performance.

- Part Four: Comprehensive Case Advancement (Chapter 13 ~ Chapter 15): This is the intersection of theory and practice. You will integrate what you have learned and build intelligent travel assistants, automated deep research agents, and even a cyber town simulating social dynamics with your own hands, tempering your construction ability in real and interesting projects.

- Part Five: Graduation Design and Future Outlook (Chapter 16): At the end of the journey, you will welcome a graduation design, building a complete multi-agent application that belongs to you, comprehensively testing your learning results. We will also look forward to the future of agents with you, exploring exciting frontier directions.

Agents are a rapidly developing field that relies heavily on practice. To achieve the best learning effect, we have provided all supporting code in the code folder of the project, and strongly recommend that you combine theory with practice. Please be sure to run, debug, and even modify every piece of code provided in the project with your own hands. You are welcome to follow Datawhale and other Agent-related communities at any time. When encountering problems, you can ask questions in the issue area of this project at any time.

Now, are you ready to enter the wonderful world of agents? Let's set off immediately!

- Video courses are being released successively (will be more detailed, practical courses will lead everyone from design ideas to implementation, giving both fish and fishing skills)

- HelloAgents framework has been updated to version V1.0.0, will continue to improve, add more useful, lightweight tools and features, compatible with learning version.

- May 2026 update agent latest learning route, click to query

- Thank you for helping reach 60k Stars! Now providing a survey for everyone to fill in the agent content they need to learn. Subsequent work "Training Agents from Scratch" will help every learner master the ability to train custom scenario agent models from zero to one.

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

- 🐛 Report Bug - If you find content or code issues, please submit an Issue

- 💡 Propose Suggestions - If you have good ideas for the project, welcome to initiate discussions

- 📝 Improve Content - Help improve the tutorial, submit your Pull Request

- ✍️ Share Practice - Share your learning notes and projects in "Community Contribution Selection"

- Chen Sizhou - Project Leader (Datawhale member, full text writing and proofreading)

- Sun Tao - Co-initiator (Datawhale member, CAMEL-AI, Chapter 9 content and proofreading)

- Jiang Shufan - Co-initiator (Datawhale member, chapter exercise design and proofreading)

- Huang Peilin - Datawhale Intentional Member (Agent Development Engineer, Chapter 5 content contributor)

- Zeng Xinmin - Agent Engineer (Nowcoder Technology, Chapter 14 case development)

- Hu Hao - Operations Organizer (Datawhale member)

- Zhu Xinzhong - Guidance Expert (Datawhale Chief Scientist - Professor at Hangzhou Institute of Artificial Intelligence, Zhejiang Normal University)

- WH (Content Contributor)

- Zhou Aojie - DW Contributor Team (Xi'an Jiaotong University, Extra02 content contribution)

- Zhang Chenxu - Individual Developer (Imperial College London, Extra03 content contribution)

- Huang Honghan - DW Contributor Team (Shenzhen University, Extra04 content contribution)

- Wang Dapeng - Datawhale Member (Senior R&D Engineer, Extra08 content contribution)

- You Yihui - Individual Developer (Nanjing University of Information Science and Technology, Extra09 content contribution)

- Yin Xin - Individual Developer (Zhejiang University, Extra10 content contribution)

- Pranav Janakiraman - Individual Developer (TinyFish, Extra11 content contribution)

- Wang Yufei - Individual Developer (Beijing University of Posts and Telecommunications, Extra12 content contribution)

- Thank you @Sm1les for your help and support for this project

- Thank you to all developers who have contributed to this project ❤️

⭐ If this project is helpful to you, please give us a Star!

If Hello-Agents is helpful to your research or work, welcome to cite:

@misc{hello_agents2025,

title = {Hello-Agents: Building an AI Agent from Scratch},

author = {Sizhou Chen and Tao Sun and Shufan Jiang and Peilin Huang and Xinmin Zeng and Xinzhong Zhu and all Hello-Agents contributors},

year = {2025},

url = {https://github.com/datawhalechina/Hello-Agents},

note = {GitHub repository}

}

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.