🧠 SlideMind AI

Transform Lecture Slides into Structured Knowledge

SlideMind AI is a production-grade, AI-powered smart learning platform that converts static lecture slides (PDF/PPTX) into dynamic, interactive study experiences — complete with summaries, topic breakdowns, AI chat tutoring, and custom practice exams.

→ Try the Live App · → View API Docs

📸 Overview

SlideMind AI takes your lecture materials and runs them through a pipeline of AI features — all accessible through a clean, focused light-theme interface built for long study sessions.

Upload once. Summarize instantly. Study smarter.

✨ Features

🤖 AI-Powered Learning

Executive Summaries — instantly distills lecture content into a concise overview with key bullet points and exam-critical focus areas
Topic Extraction — automatically identifies 3–7 distinct topics from your slides and builds a structured knowledge map
Deep-Dive Explanations — click any topic to get a teacher-style breakdown with real-world examples, formatted in clean Markdown
AI Chat Tutor — a contextual assistant that has "read" your slides and answers your specific questions 24/7
Smart Quiz Generator — generates configurable MCQ practice exams (Easy / Medium / Hard, 5–15 questions) with explanations for every answer

💾 Smart Caching

AI results (summaries, topics) are saved to the database after the first generation
Reopening a lecture from history never re-calls the AI — results load instantly at zero token cost
A ✓ AI cached badge appears on history cards so you always know what's pre-processed

📂 Document Management

Upload PDF and PowerPoint (PPTX/PPT) files up to 50MB
Full upload history per user account, sorted by most recent
One-click delete to remove any lecture from your history
Session-based document state so your active lecture persists across tabs

🔐 Authentication

Secure JWT-based signup and login
All documents and AI cache are scoped to the authenticated user
Demo mode available without login — try the full feature set with a sample lecture

🎨 UI/UX

Premium light theme (Slate/White/Brand Blue) optimized for focus
Smooth Framer Motion animations throughout
Glassmorphism cards and custom scrollbar styling
Fully responsive — works on desktop and mobile

🛠️ Technology Stack

Frontend

Technology	Purpose
Next.js 14 (App Router)	React framework with SSR
Tailwind CSS	Utility-first styling
Framer Motion	Animations and transitions
Lucide React	Icon library
React Hot Toast	Notification toasts
React Markdown	Rendering AI Markdown output
Axios	HTTP client with interceptors

Backend

Technology	Purpose
FastAPI	Async Python API framework
Google Gemini 2.5 Flash	AI summarization, chat, quiz
SQLAlchemy 2.0	Async ORM
SQLite + aiosqlite	Local database (dev)
python-pptx	PowerPoint parsing
pdfminer.six	PDF text extraction
python-jose	JWT authentication
passlib + bcrypt	Password hashing
Uvicorn	ASGI server

Infrastructure

Technology	Purpose
Google Cloud Run	Serverless container hosting
Google Cloud Build	Automated container builds
Google Artifact Registry	Container image storage
Docker	Containerization

🏗️ Architecture

┌─────────────────────────────────────────────────────────┐
│                    User's Browser                        │
│              Next.js 14 Frontend (App Router)            │
│         Tailwind CSS · Framer Motion · Axios             │
└──────────────────────┬──────────────────────────────────┘
                       │ HTTPS REST API
┌──────────────────────▼──────────────────────────────────┐
│              FastAPI Backend (Cloud Run)                  │
│                                                          │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌────────┐  │
│  │  /upload │  │  /ai     │  │  /quiz   │  │ /auth  │  │
│  └──────────┘  └──────────┘  └──────────┘  └────────┘  │
│                       │                                   │
│            ┌──────────┴──────────┐                       │
│            │                     │                        │
│     ┌──────▼──────┐    ┌────────▼────────┐              │
│     │  SQLite DB   │    │  Gemini 2.5     │              │
│     │  (Users,     │    │  Flash API      │              │
│     │  Documents,  │    │  (Summarize,    │              │
│     │  AI Cache)   │    │  Chat, Quiz)    │              │
│     └─────────────┘    └─────────────────┘              │
└─────────────────────────────────────────────────────────┘

API Endpoints

Method	Endpoint	Description	Auth
`POST`	`/api/auth/signup`	Create new account	❌
`POST`	`/api/auth/login`	Login, returns JWT	❌
`GET`	`/api/auth/me`	Get current user	✅
`POST`	`/api/upload/`	Upload and parse PDF/PPTX	✅
`GET`	`/api/upload/history`	Get user's document history	✅
`DELETE`	`/api/upload/{id}`	Delete a document	✅
`POST`	`/api/ai/summarize`	Generate AI summary (cached)	Optional
`POST`	`/api/ai/topics`	Extract topic structure (cached)	Optional
`POST`	`/api/ai/explain`	Deep-dive explanation	Optional
`POST`	`/api/ai/chat`	Chat with slide content	Optional
`POST`	`/api/quiz/generate`	Generate MCQ quiz	Optional

🚀 Getting Started (Local Development)

Prerequisites

Node.js v18+
Python v3.10+
A Google Gemini API Key

1. Clone the repository

git clone https://github.com/Ahtisham992/slidemind-ai.git
cd slidemind-ai

2. Backend setup

cd backend

# Create and activate virtual environment
python -m venv venv
# Windows:
.\venv\Scripts\activate
# Linux/Mac:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Create environment file
echo "GOOGLE_API_KEY=your_gemini_api_key_here" > .env
echo "SECRET_KEY=your_jwt_secret_here" >> .env

# Start the server
uvicorn main:app --reload

Backend runs at http://localhost:8000 · API docs at http://localhost:8000/docs

3. Frontend setup

cd frontend

# Install dependencies
npm install

# Create environment file
echo "NEXT_PUBLIC_API_URL=http://localhost:8000" > .env.local

# Start dev server
npm run dev

Frontend runs at http://localhost:3000

☁️ Cloud Deployment (Google Cloud Run)

Prerequisites

gcloud CLI installed and authenticated
A GCP project with billing enabled
Gemini API enabled: gcloud services enable generativelanguage.googleapis.com

Deploy Backend

cd backend
gcloud run deploy slidemind-api \
  --source . \
  --allow-unauthenticated \
  --region us-central1 \
  --set-env-vars GOOGLE_API_KEY=your_key,SECRET_KEY=your_secret

Deploy Frontend

cd frontend
gcloud run deploy slidemind-ui \
  --source . \
  --allow-unauthenticated \
  --region us-central1 \
  --set-env-vars NEXT_PUBLIC_API_URL=https://your-backend-url.run.app

Update environment variables (without rebuilding)

gcloud run services update slidemind-api \
  --region us-central1 \
  --update-env-vars GOOGLE_API_KEY=new_key_here

⚠️ Production Considerations

Database Persistence

The current setup uses SQLite stored inside the container, which means data is wiped on every deployment or container restart. For production, migrate to a persistent database:

Option A — Google Cloud SQL (PostgreSQL)

gcloud sql instances create slidemind-db \
  --database-version=POSTGRES_15 \
  --tier=db-f1-micro \
  --region=us-central1

Then update DATABASE_URL in backend/database.py to use postgresql+asyncpg://.

Option B — Supabase (Free tier) Sign up at supabase.com, create a project, and use the provided PostgreSQL connection string.

Token Usage & Cost

SlideMind AI uses Gemini 2.5 Flash which is highly cost-efficient:

Operation	Est. tokens	Notes
Summarize	~1,500	Cached after first call
Topic extraction	~2,000	Cached after first call
Topic explanation	~1,200	Per topic click
Quiz (10 questions)	~2,500	Per quiz attempt
Chat message	~800	Always live

At $0.15 / 1M input tokens, you can run approximately 80,000 summarizations for $1. The DB caching system ensures AI is only called once per document for summaries and topics.

📁 Project Structure

slidemind-ai/
├── backend/
│   ├── models/
│   │   ├── db_models.py        # SQLAlchemy ORM models (User, Document)
│   │   └── schemas.py          # Pydantic request/response schemas
│   ├── routers/
│   │   ├── ai.py               # AI endpoints (summarize, topics, chat, explain)
│   │   ├── auth.py             # JWT authentication endpoints
│   │   ├── quiz.py             # Quiz generation endpoint
│   │   └── upload.py           # File upload, history, delete endpoints
│   ├── services/
│   │   ├── auth_service.py     # Password hashing, JWT encode/decode
│   │   ├── file_parser.py      # PDF and PPTX parsing logic
│   │   └── gemini_service.py   # All Gemini AI API calls
│   ├── database.py             # SQLAlchemy engine and session setup
│   ├── main.py                 # FastAPI app, middleware, router registration
│   ├── requirements.txt        # Python dependencies
│   └── Dockerfile              # Container definition
│
├── frontend/
│   └── src/
│       ├── app/
│       │   ├── page.js         # Landing page with demo mode
│       │   ├── upload/         # Upload page with history and delete
│       │   ├── learn/          # Main study interface (tabs)
│       │   ├── login/          # Login page
│       │   └── signup/         # Signup page
│       ├── components/
│       │   ├── ChatAssistant.js    # AI chat sidebar
│       │   ├── SummaryPanel.js     # Summary + key points display
│       │   ├── TopicsPanel.js      # Topic cards + deep-dive view
│       │   ├── QuizPanel.js        # Interactive MCQ quiz
│       │   ├── UploadZone.js       # Drag-and-drop file upload
│       │   ├── Navbar.js           # Top navigation with auth state
│       │   └── Footer.js           # Site footer
│       ├── context/
│       │   └── AuthContext.js  # Global auth state (login, logout, user)
│       └── lib/
│           ├── api.js          # Axios API client with auth interceptor
│           └── demoData.js     # Static demo document for unauthenticated demo
│
├── docker-compose.yml          # Local multi-service development
├── DEPLOYMENT.md               # Step-by-step Cloud Run deployment guide
└── README.md                   # This file

🔧 Known Limitations & Roadmap

Current Limitations

SQLite database is ephemeral on Cloud Run (data lost on redeploy)
Quiz results are not cached or saved to history
No support for image-heavy slides (text extraction only)
Single language support (English)

Planned Features

Persistent PostgreSQL database (Cloud SQL / Supabase)
Quiz result history and score tracking
Flashcard generation mode
Multi-language support
Slide image OCR for scanned PDFs
Study session scheduling and reminders
Export summaries as PDF or Word documents
Collaborative study rooms

🤝 Contributing

Contributions are welcome! Please feel free to open an issue or submit a pull request.

Fork the repository
Create your feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

📜 License

This project is licensed under the MIT License — see the LICENSE file for details.

👨‍💻 Author

Muhammad Ahtisham

Crafted with 🧠 and ☕ by Muhammad Ahtisham

If this project helped you, consider giving it a ⭐ on GitHub!

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
backend		backend
frontend		frontend
.gitignore		.gitignore
DEPLOYMENT.md		DEPLOYMENT.md
README.md		README.md
docker-compose.yml		docker-compose.yml
logo.png		logo.png

Folders and files

Latest commit

History

Repository files navigation

🧠 SlideMind AI

📸 Overview

✨ Features

🤖 AI-Powered Learning

💾 Smart Caching

📂 Document Management

🔐 Authentication

🎨 UI/UX

🛠️ Technology Stack

Frontend

Backend

Infrastructure

🏗️ Architecture

API Endpoints

🚀 Getting Started (Local Development)

Prerequisites

1. Clone the repository

2. Backend setup

3. Frontend setup

☁️ Cloud Deployment (Google Cloud Run)

Prerequisites

Deploy Backend

Deploy Frontend

Update environment variables (without rebuilding)

⚠️ Production Considerations

Database Persistence

Token Usage & Cost

📁 Project Structure

🔧 Known Limitations & Roadmap

Current Limitations

Planned Features

🤝 Contributing

📜 License

👨‍💻 Author

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages