@gyro_ai: Most people learn about agents either staying at the conceptual level or jumping directly into frameworks, leaving the middle layer—"why design it this way"—empty. On GitHub, the book "Building Agents from Scratch" by Datawhale covers 16 chapters from basics to multi-agent systems, with its own HelloAgents framework to learn by doing…
Summary
Datawhale's open-source tutorial "Building Agents from Scratch" covers memory systems, RAG, context engineering, etc. in 16 chapters, including the HelloAgents framework, suitable for learners who want to deeply understand the internal mechanisms of agents.
View Cached Full Text
Cached at: 05/16/26, 09:16 AM
Hello-Agents
🤖 Building Agents from Scratch
From foundational theory to practical application, master the design and implementation of agent systems.
Scan the QR code to provide feedback and help the new project grow together.
⭐ If this project helps you, please give us a Star!
Scan the QR code to join the reader discussion group and exchange ideas with more learners.
Scan the QR code to follow Datawhale’s official account and get more high-quality open-source content.
Similar Articles
datawhalechina/hello-agents
Datawhale社区发布的开源中文教程《从零开始构建智能体》，系统性讲解AI原生智能体的理论与实践，涵盖从基础原理到自研框架HelloAgents的完整学习路径。
@xiaojianjian567: Introduction to the Excellent GitHub Project with 45k Stars: hello-agents. Produced by Datawhale, it gained 1,200+ stars in 7 days and is surging. What does this project do? One-sentence description: Teaches you from scratch how to build real AI Agents—not just by calling APIs, but by understanding core principles, system design, and…
Datawhale's open-source project, hello-agents, is gaining rapid popularity on GitHub. It aims to systematically teach developers the core principles, multi-agent collaboration, and system design required to build AI Agents from scratch, going beyond mere API usage.
@vintcessun: Tonight I came across a learning roadmap project that redefined where to start learning Agent. I used to think Agent was just a pile of tools and frameworks, but its core is the "observe-think-execute" loop and the harness engineering's organization of permissions, state, and backtracking. It breaks down learning into building a minimal Agent loop from scratch all the way to deploying a real Agent, with 8 stages, each with clear deliverables and recommended resources — not just links but an actionable todo list. This systematic approach made me realize my previous learning was too fragmented.
An open-source learning roadmap project called Agent-Learning-Hub, which breaks down AI Agent learning into 8 stages from building a minimal Agent loop to production deployment, providing executable todo lists and recommended resources, maintained by members of the Datawhale community.
@Xudong07452910: Open-Source Project Recommendation: Agent-Learning-Hub — A Systematic Learning Roadmap and Resource Repository for Building Reliable AI Agents
Agent-Learning-Hub by DataWhale is a systematic learning roadmap and resource repository for AI Agent development, featuring 8 learning stages and 11 progressive projects to help developers build reliable AI agents from scratch. It has gained 3.9k+ stars.
@XAMTO_AI: Guys, still learning AI Agent in bits and pieces? Check this out! A 22-chapter skeleton course on building a production-grade Agent system from 0 to 1. Tool calling, Agent loop, memory system, multi-Agent collaboration, human-machine interaction, backend architecture, observability, cost optimization, security... The coolest part is the core gameplay — using Agent to teach Agent. The author only provides the skeleton, and the AI assistant fills in the code and details. Learn by doing. This is the learning style for 2026; the era of rote memorization is over.
A 22-chapter skeleton course on building production AI agents, using an innovative approach where the AI partner fills in details. The course covers tool calling, agent loops, memory, multi-agent collaboration, and more.