---
title: LearnHub AI Platform
emoji: 🎓
colorFrom: purple
colorTo: blue
sdk: docker
app_port: 7860
pinned: false
---

# LearnHub - Phase 2 AI Learning Platform

**Advanced AI-powered learning platform with multi-user support, course management, and RAG-based tutoring.**

## Features

✅ **User Authentication** - JWT-based auth with student/instructor/admin roles  
✅ **Course Management** - Modules, lessons, quizzes, and progress tracking  
✅ **AI Tutor** - RAG-powered Q&A using PostgreSQL pgvector + Google Gemini  
✅ **Smart Features** - Auto-generate summaries, MCQs, and progressive hints  
✅ **Forum** - Discussion boards for students and instructors  
✅ **Vector Search** - pgvector for semantic search over course materials

## Tech Stack

- **Backend**: FastAPI + SQLAlchemy + PostgreSQL  
- **Vector DB**: pgvector extension for PostgreSQL  
- **AI**: Google Gemini 2.5 Flash + LangChain  
- **Auth**: JWT tokens with bcrypt password hashing  

## Deployment on HuggingFace Spaces

This Space requires a **PostgreSQL database with pgvector extension**.

### Option 1: Use Supabase (Recommended)

1. Create a Supabase project at https://supabase.com
2. Go to Project Settings → Database → Connection string (URI)
3. Enable pgvector: Go to SQL Editor and run:
   ```sql
   CREATE EXTENSION vector;
   ```
4. Add your Supabase connection string as `DATABASE_URL` secret in HF Spaces

### Option 2: Use Render PostgreSQL

1. Create a PostgreSQL database on Render.com
2. Connect and enable pgvector:
   ```sql
   CREATE EXTENSION vector;
   ```
3. Use the external connection string as `DATABASE_URL`

### Required Secrets

Configure in **Settings → Variables and secrets**:

```
DATABASE_URL=postgresql+asyncpg://user:password@host:5432/database
GOOGLE_API_KEY=your_google_api_key
JWT_SECRET_KEY=your-super-secret-jwt-key
CORS_ORIGINS=https://your-frontend.vercel.app
```

### Run Database Migrations

After deployment, you need to run Alembic migrations:

```bash
# Connect to your database and enable pgvector
CREATE EXTENSION IF NOT EXISTS vector;

# Or use the migration script in alembic/versions/
```

## API Endpoints

| Endpoint | Description |
|----------|-------------|
| `POST /auth/register` | Register new user |
| `POST /auth/login` | Login and get JWT token |
| `GET /courses` | List all courses |
| `POST /tutor/ask` | Ask AI tutor a question (RAG) |
| `POST /tutor/generate-mcq` | Generate MCQ questions |
| `POST /tutor/hint` | Get progressive hints |
| `GET /docs` | Interactive API documentation |

## Local Development

```bash
# Setup
cd backend
pip install -r requirements.txt

# Configure .env
cp .env.example .env
# Edit .env with your credentials

# Run migrations
alembic upgrade head

# Start server
uvicorn app.main:app --reload --port 8001
```

## Differences from Phase 1

| Feature | Phase 1 | Phase 2 |
|---------|---------|---------|
| Database | SQLite | PostgreSQL |
| Vectors | Pinecone | pgvector (PostgreSQL) |
| Auth | None | JWT multi-user |
| Courses | No | Full course management |
| AI Features | Basic Q&A | Q&A + MCQ + Summaries + Hints |
| Users | Single | Multi-user with roles |

---

**Note**: This is the backend API only. For a full app experience, deploy the Next.js frontend separately and connect via CORS.