ai-shifu/ai-shifu

AI‑Shifu

What it is – AI‑Shifu is an open‑source platform that lets creators, instructors, or corporate training teams turn a single lesson outline into a personalized, interactive AI tutor for each learner. It runs on a large‑language‑model (LLM) backend and provides real‑time explanations, probing questions, assessments, and feedback, all wrapped in a web UI that can be embedded in websites or learning portals.

Key capabilities

• Personalized explanation engine – Generates learning paths, tone, and depth based on a learner’s background, goals and skill level.
• Interactive Q&A & probing – Breaks down learner questions, asks clarifying follow‑ups, and suggests next steps during a session.
• Rapid course assembly – Authors supply a high‑level framework; the system expands it into full lessons, activities, and quizzes.
• One‑click deployment – Docker‑compose files (latest or pinned) let you spin up the whole stack (MySQL, optional Redis, API, and web front‑end) in minutes.
• Multi‑channel integration – Can be embedded in corporate portals, public websites, or dedicated training platforms.
• Extensible TTS – Optional text‑to‑speech via Volcengine or other providers.

Typical users

• Course creators who want to reuse a single syllabus across many learners without writing individualized content.
• Enterprise training teams that need role‑specific learning paths generated from a common knowledge base.
• Educators looking for a AI‑driven coaching assistant that can answer student questions and generate assessments on the fly.

How to try it

1. Hosted demo – Visit https://ai-shifu.com to experience a ready‑made AI‑guided course.
2. Self‑host – Clone the repo and start the Docker stack:

   git clone https://github.com/ai-shifu/ai-shifu.git
   cd ai-shifu/docker
   cp .env.example.full .env   # edit .env
   # set your LLM provider API key and LLM_MODEL_1_ID (e.g., an OpenAI model ID)
   docker compose -f docker-compose.latest.yml up -d
   

   After the containers are up, open http://localhost:8080. Use any phone number to log in; the default demo verification code is 1024 (change for production). The first verified user becomes admin/creator.

Development workflow – dev_in_docker.sh builds backend and frontend images from the current source and launches a hot‑reload compose file, making it easy to iterate on code without installing Python/Node locally.

Roadmap highlights

• AI agent for rapid script generation and maintenance
• Centralised knowledge‑base integration
• Speech input and output (voice‑first interaction)

Where to find details

• Installation guide: INSTALL_MANUAL.md
• Docker image publishing: GitHub Container Registry (ghcr.io/ai-shifu/...)
• Internationalisation: src/i18n/ and docs in docs/references/i18n.md
• TTS configuration: set VOLCENGINE_TTS_* env vars and select volcengine_http in the UI.

Bottom line – AI‑Shifu is a full‑stack, containerised AI tutoring system that turns static teaching material into a dynamic, personalized learning experience, aimed at educators and enterprises that want to scale one‑on‑one instruction without writing bespoke content for every learner.