easy-vibe README

The project's README from github.com/datawhalechina/easy-vibe. As one markdown file: README.md.

Easy-Vibe Logo Easy-Vibe — Think it. Build it. From first idea to real product.

Learn AI coding from zero by shipping real products.
从零开始学 AI 编程,把想法真正做成产品。

datawhalechina/easy-vibe | Trendshift

你好 · Hello · 哈囉 · こんにちは · 안녕하세요 · Hola · Bonjour · Hallo · مرحبا · Xin chào
Our tutorial supports 10 languages. Let's code together!
我们的教程支持 10 种语言,欢迎世界各地的朋友一起 coding!

🚀 Start Exploring · ✨ Interactive Tutorial · 🦞 Learn OpenClaw · 📖 Table of Contents
🚀 开始体验 · ✨ 交互式教程 · 🦞 学习 OpenClaw · 📖 查看目录

Read Online · Learning Map
开始阅读 · 学习地图

Stars Forks License

English 简体中文 繁體中文 日本語 Español Français 한국어 العربية Tiếng_Việt Deutsch


A beginner-friendly learning map
Clear guidance from zero, so you can stop "learning and forgetting"

Step-by-step visual tutorials
Detailed walkthroughs that feel like learning with a private tutor

Immersive simulated coding
Virtual mouse guidance helps you quickly learn the core IDE workflow

Visible AI principles
Animated explanations make it easy to see how AI generates images

Learn RAG like a game
Interactive components let you click through the full RAG data flow

Visual terminal concepts
Command-line behavior becomes intuitive when the underlying logic is visualized

⭐ Star the repo here to help accelerate updates ❤️

Share Your Vibe Story

📝 Have your own vibe coding story? Submit it here and inspire others!

Table of Contents

Why Easy-Vibe

Want an expense tracker? Say it.

Need a booking system with WeChat login? Say it.

Want a blog with comments? Say it.

In the AI era, programming starts by describing what you want.

Easy-Vibe teaches you how to turn that into a real product.

🔥 News

Past News
  • [2026-03-02] 🦞 OpenClaw and AI Agent friendly support: Added llms.txt so OpenClaw, Claude, Cursor, Trae, and other AI agents can quickly understand the repository structure and find the right tutorial content.
  • [2026-03-01] The Advanced Development section has been comprehensively upgraded with deep guides for Claude Code, including MCP, Skills, Agent Teams, and more, along with eight cross-platform project tutorials.
  • [2026-02-25] Updated the Appendix Knowledge Base, now covering 9 knowledge areas and 80+ interactive topics.
  • [2026-01-27] Added Android and iOS app development tutorials.
  • [2026-01-19] Released interactive demos for Prompt Engineering, AI history, authentication design, Git principles, and more.
  • [2026-01-16] Reorganized the project structure and formally established a beginner entry path.
  • [2026-01-14] Completed a large update to the Stage 1 product prototyping docs.
  • [2026-01-13] Refactored the documentation architecture and fully enabled multi-language support.
  • [2026-01-01] Released the core learning map for the project.

Who This Is For

Your Learning Paths

🎮 I want a fast first win

Best for: Everyone What you will learn: What AI coding actually feels like through a simple, concrete hands-on example What you will get: A clear first impression of vibe coding and how to work with AI by conversation

Start here

💡 I want to turn an idea into a product prototype

Best for: Beginners / product managers / founders What you will learn: Learning roadmap, AI IDE tools, idea validation, prototyping, AI capability integration, and full demo iteration What you will get: A demoable AI product prototype you can actually show to users or teammates

Start learning

🚀 I want to build full-stack products end to end

Best for: Junior developers / indie hackers / advanced learners What you will learn: Frontend workflows, design-to-code, databases, backend APIs, deployment, billing, and major projects What you will get: The ability to independently ship modern AI-enabled web applications

Start learning

AI-Native: I want advanced Claude Code and agent workflows

Best for: Developers interested in AI-native engineering What you will learn: Claude Code, MCP, Skills, Agent Teams, long-running tasks, Spec Coding, and cross-platform app delivery What you will get: A stronger workflow for complex AI-assisted development and automation

Go to advanced development

📚 I want reference material and fundamentals

Best for: Everyone What you will learn: Computer fundamentals, frontend/backend basics, infrastructure, AI principles, and engineering practices What you will get: A long-term reference knowledge base covering 9 major knowledge areas

Browse the knowledge base

Study Suggestions

📖 Content Navigation

Easy Vibe learning map: from idea validation to full-stack delivery and advanced AI coding

I. Beginner Entry

Section Key Content
Learning Map A guided overview of the full learning journey
AI Era: If You Can Speak, You Can Code Get your first feel for AI coding through examples like Snake
Master AI Programming Tools Learn how AI IDE tools work and build simple local projects with them
Find Great Ideas Learn how to discover and validate product ideas worth building
Build Product Prototypes Move from requirements to single-page and multi-page product prototypes
Integrate AI Capabilities Integrate text, image, and video AI features
Complete project practice Simulate real scenarios, collect user feedback, and iterate on a full project

Appendix: Product and Business Thinking

Section Key Content
Product Thinking and Solution Design Core frameworks for going from zero to one with a product
AI Application Scenario Reference (B2B & B2C) B2B industry and B2C consumer AI application scenarios

Appendix: User Research and Requirement Validation

Section Key Content
Where to find ideas: 3 reference sources that work best for beginners Build a reliable pipeline for finding concrete product opportunities
Double Diamond: first do the right thing, then do it right Use a structured process to move from scattered inspiration to a workable direction
Use Jobs to Be Done to find what users really want done Analyze user goals through real tasks instead of surface-level feature requests
The Mom Test: a user interview method for validating demand Learn how to ask better questions and avoid false-positive feedback

Appendix: Technical Solutions

Section Key Content
What to do if you encounter errors Common vibe coding issues and how to troubleshoot them
Comparison of Seven AI Programming Tools Compare major AI coding platforms through hands-on testing
Design Websites with Agents Learn multi-agent collaboration in practice

II. Junior and Mid-Level Developers

Frontend

Section Key Content
Frontend 0: Build Your Own Asset-Production Agent with Lovart Use Nanobanana and Lovart to batch-generate visual assets and build a drawing agent with intent recognition
Frontend 1: Figma & MasterGo Basics Learn the workflow from design drafts to implementation-ready UI thinking
Frontend 2: Build Your First Modern App - UI Design Learn the UI design foundations behind modern application interfaces
Frontend 3: UI Guidelines and Multi-Product Design Improve consistency and aesthetics across multiple products with shared UI rules
Frontend 4: Make Interfaces Beautiful with LLMs and Skills Use prompts and plugins to make AI generate more polished, distinctive interfaces
Frontend 4: Let's Build Hogwarts Portraits Build an interactive AI-image frontend project from scratch
Frontend 6: From Design Prototype to Project Code Turn design prototypes into frontend code that can really run in the browser
[Frontend 7: Upgrade Your UI with Modern Component Libraries](https://datawhalechina.github.io/easy-vibe/en/stage

easy-vibe (datawhalechina/easy-vibe)

Searching Database…

19.8K ·GitHub
这是什么东西?Easy-Vibe 是 Datawhale 出品的一门面向初学者的课程('vibe coding 101'),教你用 AI 的方式做产品:通过真正上线作品来学编程,从一个初步想法一路做到能跑起来的应用。
这是需要安装的应用,还是在浏览器里阅读的课程?是在浏览器里阅读的课程:一个可视化的学习地图,提供一步步的图文讲解和互动演示,用 VitePress 搭建。也可以在本地运行,需要 Node 18+('npm install && npm run dev';仓库里还附带 Dockerfile)。
开始之前我需要先自己准备好 AI 编程工具吗?不需要——课程会从零教你使用各种 AI IDE 工具,并包含了七个 AI 编程平台的实操对比。第 3 阶段会深入讲解 Claude Code(MCP、Skills、Agent Teams),仓库还提供 llms.txt,方便 OpenClaw、Claude、Cursor、Trae 这类智能体浏览。
我不是开发者,也能跟着学吗?可以——第 1 阶段就是为零基础新手、第一次做产品的人和独立开发者设计的;产品经理、创业者和学生也是明确的目标人群,第 2-3 阶段则面向从初级到高级的开发者。
学的过程中我实际会做出什么?你会先做一个简单的贪吃蛇游戏来找找 AI 编程的感觉,然后做出单页和多页原型、一个 SaaS 毕业项目(带 Stripe 支付的文案生成网站)、一个 AI 图像前端项目(Hogwarts Portraits),进阶路线里还有 Android/iOS 应用。
课程是怎么做到互动的?演示包括:模拟编码——用一只虚拟鼠标带你走完核心的 IDE 操作流程;一个可以点击闯关的游戏,展示完整的 RAG 数据流;用动画讲解 AI 如何生成图像;还有可视化的终端行为演示。
Vibe Stories 是什么?这是首页的一个板块,收录了真实用户的故事:一位乡村小学老师、一名大学生、一位高中信息技术老师和一位卡车司机,他们都用 AI 做出了真实的产品。
内容还在持续维护吗?是的——2026 年的更新重组了第 1 阶段,围绕发现和验证真实需求展开,新增了带 Stripe 支付的 SaaS 毕业项目,以及一个用户调研附录,涵盖双钻模型、Jobs to Be Done 和 The Mom Test。
Size
537.4 MB
Stars
19.8K
Age
9mo
Commits · 90 Days