Menu Buddy

• 

  Main screen

• 

  The picture used for the menu

• 

  it now extract the text and search for corresponding items

• 

  the result

• 

  and even more items read

Inspiration

I often found myself at restaurants, staring at menus filled with unfamiliar dish names and no pictures. It’s tough to decide what to order when you have no idea what the food looks like. This common frustration inspired me to build an app that makes dining out a more visual and informed experience.

This could even help foreign travellers order at the restaurant by saying the name of the dish on a button press in the language of the menu.

How It Was Built

The project was built using:

• Fully vibe coded using windsurf.
• Web developed using next.js and tailwind.
• A backend service or API integration to process the taken picture with Open AI and fetch representative images of the dishes based on the recognized text using Tavily.
• A clean, user-friendly UI/UX that allows users to easily snap a photo, see dish names, and view corresponding images.

Challenges Faced

Some of the key challenges I encountered:

• Dealing with the different API and process the data properly.
• Ensuring the app is easy to use.

Next Step

The next step would be to create a mobile app for people to quickly use with their phone. It could also allow restaurant owner a way to help them create a more visual menu or to setup their, for example, ship the dish menu. A calorie counter and allergies warning would be great addition next.

Built With

• nextjs
• openai
• react
• tailwind
• tavily
• typescript
• windsurf