Decor.AI

Inspiration

As a professional in IT, I find myself loving beautifully decorated spaces but lacking both the taste and patience to select furniture. However, I still appreciate well-decorated places. That said, I think it's still expensive to hire a professional designer, which is why I believe it's important to democratize access to beautiful, well-decorated spaces - making professional-quality design accessible to everyone at an affordable price.

What it does

Decor.AI allows users to upload a photo of their room and instantly transform it using AI. Users can select their preferred space type (living room, bedroom, office, etc.), choose color palettes, set a budget, and optionally specify preferred stores. The app generates stunning AI-powered redesigns maintaining the original room layout while replacing furniture and decor to match the desired style. Additionally, it creates a complete shopping list with real product links and prices to help users purchase exactly what they need.

How we built it

The application is built with React 19 and TypeScript, using Vite as the build tool. We leverage Google's Gemini AI to generate both the image redesigns and intelligent shopping lists. The frontend uses Tailwind CSS for styling and includes components for file upload, option selection, budget setting, and results display. The backend service integrates with Gemini API for image generation and text analysis, using modern web standards like import maps for efficient loading.

Challenges we ran into

One of the main challenges was configuring environment variables properly in Vite for the LLM API integration. We initially had issues with the blank screen because the API key wasn't being loaded correctly. Additionally, working with multiple response modalities from LLM (image + text) required careful handling of the API responses. Ensuring proper TypeScript types across the application and managing file uploads with base64 conversion also presented some technical hurdles.

Accomplishments that we're proud of

We're particularly proud of creating a user-friendly interface that makes professional interior design accessible to everyone. The app successfully combines cutting-edge AI technology with practical e-commerce integration, providing users not just inspiration but also actionable shopping lists. We've built a complete end-to-end solution that can run both locally and deploy to cloud infrastructure.

What we learned

Through this project, we gained hands-on experience with modern AI APIs, particularly Google's Gemini multimodal capabilities. We learned about the importance of proper environment variable configuration in frontend applications and the challenges of integrating AI services into web apps. We also discovered the value of combining AI-generated content with practical business logic like shopping list generation.

What's next for Decor.AI

We're planning to add more customization options like multiple style variations in each generation, furniture style filtering, and improved budget optimization. We'd also like to implement user accounts for saving designs, social sharing features, and potentially a marketplace for user-generated designs. Future versions may include AR integration for visualizing designs in real spaces and partnerships with furniture manufacturers for exclusive deals.

Built With

• css
• gemini
• node.js
• react-19
• tailwind
• typescript
• vite