GitHub radar
DeepTutor: A Free AI Education Platform
A local dockerized AI powered learning environment which allows you to convert your pdf/youtube videos into your own courses with quizzes and an AI tutor that keeps track of your learning progress (3 layers of memories: trace, summary, synthesis)
DeepTutor is an open source AI based learning system built by HKUDS team at Hong Kong University from Dec 2025. Users can upload their pdf/books/youtube urls to build an interactive learning system including quizzes/flashcards/progress tracking. It has an AI tutor which supports 3 level memories(trace, summary, synthesis) to remember our learning progress during multiple visits. Users can also integrate with live ai agents like Claude Code, Codex, DeepSeek, Kimi etc., and talk with them in chat mode. It supports many RAG engines like LlamaIndex, PageIndex, GraphRAG, LightRAG. Users can write markdown notes and import Obsidian vault. Enjoy immersive youtube watching and time based tutoring. All the functionalities can be easily deployed with only one docker image(ghcr.io/hkuds/deeptutor:latest) or pip install deeptutor. It’s under active development(v1.6.8 released on Sep 14, 2026).
Why a vibe-coder should care
Sometimes when we are coding in Vibe, we need to learn some knowledge quickly by asking questions and doing some tests rather than reading all the documentations. Now with DeepTutor, we can organize our own documents to be a learning system with an AI tutor which remembers our learning progress. One single docker command, we can have Claude Code as our learning partner without any infra setup.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Run DeepTutor from https://github.com/HKUDS/DeepTutor with: docker run --rm -p 3782:3782 ghcr.io/hkuds/deeptutor:latest — open http://localhost:3782 and help me connect an AI model. Ask me for keys if needed.
Runs on any laptop — only Docker required.
Open on GitHub▌ More finds