An autonomous AI agent that transforms empty rooms into beautifully designed spaces using GPT-4.1-mini and Arcade Google Shopping SDK.

• Autonomous Design Process: Upload a photo of an empty room and let the AI handle everything
• Room Analysis: Uses GPT-4.1-mini with vision capabilities to analyze room dimensions, lighting, and features
• Smart Planning: Creates detailed interior design plans with furniture recommendations
• Furniture Shopping: Searches real products using Google Shopping via Arcade API
• Design Generation: Creates detailed descriptions of the final room design
• Real-time Progress: Track the agent's progress through each step
• Markdown Reports: Get detailed design plans in well-structured markdown

This app uses FastAPI backend which works best on platforms that support full Python applications.

1. Sign up at railway.app

2. Install Railway CLI:

   npm install -g @railway/cli

3. Deploy:

   railway login
   railway link
   railway up

4. Add environment variables in Railway dashboard:

   • OPENAI_API_KEY
   • ARCADE_API_KEY
   • USER_ID

5. Deploy frontend separately on Vercel:

   cd frontend
   vercel

1. Sign up at render.com

2. Connect your GitHub repo

3. Create new Web Service with:

   • Runtime: Python
   • Build: pip install -r requirements.txt
   • Start: cd backend && uvicorn main:app --host 0.0.0.0 --port $PORT

4. Add environment variables in Render dashboard

1. Install Heroku CLI and login:

   heroku login

2. Create app and deploy:

   heroku create your-app-name
   git push heroku main

3. Set environment variables:

   heroku config:set OPENAI_API_KEY=your_key
   heroku config:set ARCADE_API_KEY=your_key
   heroku config:set USER_ID=your_id

1. Create Dockerfile (we can add this if needed)
2. Deploy to:
   • Google Cloud Run
   • AWS App Runner
   • Azure Container Apps
   • DigitalOcean App Platform

The frontend can be deployed separately on:

• Vercel (recommended): cd frontend && vercel
• Netlify: cd frontend && netlify deploy
• GitHub Pages: With GitHub Actions

• Update API_URL in frontend after backend deployment
• Consider using a database for production (PostgreSQL, MongoDB)
• Use cloud storage for images (S3, Cloudinary)
• Set up proper CORS origins for production

• Core Logic: All business logic and AI processing
• Agent System: Autonomous agent with state management
• API Integration: OpenAI responses API and Arcade Google Shopping
• Session Management: Handles multiple concurrent design sessions
• File Handling: Room image uploads and storage

• Pure Presentation: No business logic, only UI
• Real-time Updates: Polls backend for agent status
• Type Safety: Full TypeScript implementation
• Responsive Design: Works on desktop and mobile

• Python 3.8+
• Node.js 16+
• OpenAI API key
• Arcade API key and user ID

1. Navigate to the backend directory:

cd backend

2. Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file based on env.example:

cp env.example .env

5. Add your API keys to .env:

OPENAI_API_KEY=your_openai_api_key_here
ARCADE_API_KEY=your_arcade_api_key_here
ARCADE_USER_ID=your_arcade_user_id_here

1. Navigate to the frontend directory:

cd frontend

2. Install dependencies:

npm install

From the project root directory:

On macOS/Linux:

./start.sh

On Windows:

start.bat

This will start both the backend and frontend servers automatically.

In the backend directory:

python main.py

The API will be available at http://localhost:8000

In the frontend directory:

npm run dev

The frontend will be available at http://localhost:5173

1. Open the application in your browser
2. Upload a photo of an empty room
3. The AI agent will:
   • Analyze the room characteristics
   • Create a design plan
   • Search for suitable furniture
   • Generate the final design
4. Monitor progress in real-time
5. View the complete design plan and results

1. Room Analysis 🔍

   • Identifies room type and dimensions
   • Analyzes lighting conditions
   • Detects existing features

2. Design Planning 📝

   • Selects appropriate design style
   • Creates furniture list
   • Plans room layout

3. Furniture Shopping 🛍️

   • Searches Google Shopping for each item
   • Collects product images and prices
   • Builds shopping list

4. Design Generation 🎨

   • Creates final room visualization
   • Provides cost estimates
   • Generates design description

• POST /api/upload - Upload room image
• POST /api/agent/start/{session_id} - Start design agent
• GET /api/agent/status/{session_id} - Get agent status
• GET /api/agent/plan/{session_id} - Get design plan
• GET /api/agent/results/{session_id} - Get final results
• DELETE /api/agent/{session_id} - Stop and cleanup session

The agent behavior can be customized in backend/agent.py:

• Furniture search limits
• Design styles
• Budget constraints
• Room types

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Open a pull request

This project is licensed under the MIT License.

• OpenAI for GPT-4.1-mini
• Arcade for Google Shopping SDK
• React and FastAPI communities