Yupi AI Quiz Learning Mini-Program (yu-ai-learn)
A teaching project that builds a WeChat mini-program from scratch with Taro, React, FastAPI, LangChain, LangGraph and DeepSeek, where a single sentence produces AI-generated learning quizzes, with a retrieval-augmented mode for private documents and a full video and text course.
This is a tutorial project that walks through building a WeChat mini-program end to end using an AI-assisted development workflow, covering requirements research through deployment. The finished app turns a one-sentence learning goal into an AI-generated quiz: the user picks a topic, the system searches the web for current material through a LangGraph ReAct agent using the Tavily search API, generates multiple-choice and true-or-false questions, grades answers immediately with an explanation, and produces a final review report with a mastery score and weak points.
It also supports uploading private PDF, Word, Markdown or plain-text documents, which are parsed, chunked, embedded and stored in a Chroma vector database so quizzes can be generated from that material instead. Other features include AI-generated illustrations for each question through a text-to-image model stored on Tencent Cloud object storage, and WeChat one-tap sign-in with an optional guest mode.
The stack is Taro and React for the frontend and Python FastAPI, LangChain and LangGraph for the backend, using the DeepSeek model, MySQL and JWT authentication, deployed with Docker to WeChat Cloud Hosting. The project ships with a full video course, a written tutorial, and interview-preparation material.