Hands-on AI projects from beginner to expert. Pick a challenge and start building.
Choose an appropriate learning topic and generate an interactive educational website
Design and implement a multi-agent pipeline consisting of a 'Researcher', 'Writer', and 'Reviewer' that work together to complete the task of generating a high-quality industry report.
Using LangChain's ReAct (Reasoning + Acting) Agent framework, build an intelligent Agent that can receive high-level goals, autonomously decompose them into subtasks, formulate execution plans, and invoke multiple tools to progressively complete tasks. You will need to equip the Agent with tools such as search, calculation, file read/write, and implement full chain-of-thought tracking and task state management.
Build an intelligent Q&A system that supports PDF/text document uploads using LangChain's Document Loaders, text splitters, vector databases, and retrieval chains. Once a user uploads a document, the system can accurately answer questions based on the document's content rather than relying on the model's own training knowledge.
Build a personal assistant with long-term memory using LangChain's Memory module. The assistant should be able to remember the user's name, preferences, and conversation history, and proactively leverage these memories in future interactions to deliver a personalized experience.
Integrate multiple external tools — such as a search engine, calculator, and weather lookup — into a LangChain Agent, enabling it to automatically select and invoke the right tool based on user questions. Through this challenge, you'll master how to define, register, and use LangChain tools, ultimately building an intelligent Agent with true "action capabilities".
Using LangChain's ChatModel and PromptTemplate, build a command-line chatbot that supports multi-turn conversations from scratch. You'll learn how to combine an LLM, prompt templates, and conversation memory into a complete Chain, gaining a solid understanding of LangChain's core architecture.
用 AI Agent 重新定义,你的工作方式。 8 个真实用户案例，8 种 AI Agent 配置方案。 选择你的角色，跟随配置指南，亲手搭建属于自己的 OpenClaw 自动化工作流。 你也可以按照你自己的需求来配置你自己的超级 AI 团队
Build the classic Snake game with smooth movement, power-ups (speed boost, score multiplier), increasing difficulty, a local leaderboard, and retro-styled graphics.
Build a responsive portfolio website with dark/light mode toggle, smooth scroll navigation, project showcase section, and contact form. Use modern CSS techniques like grid, flexbox, and animations.
Build a full-stack URL shortener with custom short links, click analytics (geographic, device, referrer), QR code generation, and a dashboard to manage all links.
Create a tool that analyzes a project's structure, dependencies, and technology stack, then generates optimized CI/CD pipeline configurations for GitHub Actions, GitLab CI, or Jenkins with caching, testing, and deployment stages.
Create a tool that translates spoken dialogue in a video into another language and generates dubbed audio with voice cloning. Preserve the original speaker's tone and timing while replacing the audio track with translated speech.
Build a beautiful chat UI for AI conversations with streaming response display, message history, conversation management, code block rendering, and markdown support.
Build an all-in-one AI-powered developer platform combining code generation, review, testing, documentation, deployment, and monitoring. Include team collaboration, project management, and a plugin system for extending functionality.
Build a browser-based code editor with an AI pair programmer that provides real-time inline suggestions, multi-file context awareness, intelligent autocomplete, integrated terminal, and conversational coding assistance in a side panel.
Build a full documentation platform that auto-generates docs from codebases, keeps them in sync with code changes, supports interactive examples, versioning, and uses AI to answer developer questions about the documented API.
Build an AI-powered architecture advisor that analyzes a codebase or project requirements and recommends system architecture, design patterns, technology choices, and scalability strategies with visual diagrams.
Build a semantic code search engine that understands what code does, not just pattern matching. Index a codebase using embeddings so developers can search with natural language queries like 'function that validates email addresses'.
Build a tool that generates complete full-stack applications from a high-level description. Produce a Next.js or similar project with database schema, API routes, UI components, authentication, and deployment configuration.
Create a tool that generates comprehensive test scenarios from user stories, feature specs, or existing code. Produce end-to-end test scripts, integration test plans, and acceptance criteria with Playwright or Cypress code.
Build a tool that generates React, Vue, or Svelte components from natural language descriptions or wireframe sketches. Include props, state management, styling, accessibility attributes, and Storybook stories.
Create a tool that assists with code migration between frameworks, languages, or API versions. Translate React class components to hooks, JavaScript to TypeScript, REST to GraphQL, or migrate between major framework versions.
Build a visual database schema designer powered by AI. Users describe their data requirements in natural language, and the tool generates an entity-relationship diagram, SQL DDL, and ORM model files with relationship mapping.
Create a tool that takes an API specification (OpenAPI, GraphQL schema, or natural language description) and generates a complete, working API with routes, controllers, validation, and database models.
Build an interactive debugging assistant that helps developers diagnose issues. Users describe the bug or paste error output, and the AI asks clarifying questions, suggests hypotheses, and walks through debugging steps.
Create a tool that identifies code smells and suggests refactoring opportunities. Detect long methods, duplicated code, complex conditionals, and deep nesting, then generate refactored versions with explanations.
Build an AI-powered code review tool that analyzes pull requests or code submissions for bugs, security vulnerabilities, performance issues, and style violations. Present findings as inline annotations with severity levels.
