Inspiration

I personally struggled with decorating and improving my room — I didn’t know where to start or what to buy. I watched YouTube videos, but they showed ideal setups that weren’t practical for my space or budget. Since I had a limited budget, most online advice didn’t apply to me. I wanted something quick and simple that could tell me exactly how to improve my room and what to buy — instantly and affordably.

What it does

You take a photo of your room, choose your style and budget, and the app analyzes your space to generate a makeover plan. It finds the best products you can buy within your budget and shows a visual comparison along with direct links to platforms like Amazon and Walmart so you can instantly purchase what you need.

How we built it

We built the app using Next.js. We used Google and OpenAI APIs to analyze room images and understand user preferences. Then we integrated SerpAPI to fetch real, purchasable products so users can directly buy items and bring the design to life.

Challenges we ran into

We explored multiple approaches such as 3D room modeling and AR-based placement to make products movable in the space. However, these approaches added complexity. We had to iterate through different ideas before finding the right balance between usability, speed, and practicality.

Accomplishments that we're proud of

We tested the app on various room types, including my own personal room, and it consistently generated useful and practical recommendations. We’re proud that it not only suggests improvements but also provides exact items users can buy to achieve the transformation.

What we learned

We learned the importance of rapid iteration. Many ideas sound great in theory, but only through implementation and testing can you determine what truly works. Finding the right balance between ambition and simplicity was key.

What's next for Roomify

We plan to add authentication, enable users to purchase directly within the app, and create instant carts so users can seamlessly go from inspiration to checkout in one flow.

Built With

  • dall-e-3
  • google-gemini-2.5-flash
  • gpt-image-1)
  • next.js-16
  • openai-api-(gpt-4o-vision
  • react-19
  • serpapi
  • tailwind-css-v4
  • typescript
  • vercel
Share this project:

Updates