-
scan page
Inspiration
Millions of people live with food allergies or dietary restrictions, yet restaurant menus are often unclear or incomplete. We wanted to create an AI-powered tool that helps people quickly understand potential risks before ordering, making dining out easier and more inclusive.
What it does
MenuSafe uses AI to analyze restaurant menu items or ingredient lists for allergies and dietary restrictions. Users can paste menu text or scan a menu, select personalized dietary profiles, and instantly receive a clear safety assessment, confidence level, explanations, and suggested questions to ask restaurant staff.
How we built it
We built MenuSafe using React Native with Expo for the frontend and integrated an LLM API for intelligent menu analysis. The app uses structured prompts to return consistent JSON responses, which are displayed in a clean interface with personalized results for different users and dietary needs.
Challenges we ran into
Our biggest challenges were handling AI responses reliably, designing an intuitive interface for multiple dietary profiles, and resolving dependency and Expo SDK compatibility issues while preparing for the hackathon deadline.
Accomplishments that we're proud of
We're proud of creating a polished AI-powered prototype that provides personalized menu analysis, supports multiple users with different dietary restrictions, explains its reasoning instead of giving simple yes/no answers, and generates helpful questions for restaurant staff.
What we learned
We learned how to integrate large language models into a real-world application, design reliable AI workflows with structured outputs, collaborate effectively under time pressure, and balance technical complexity with a user-friendly experience.
What's next for MenuSafe
We plan to add OCR for live menu scanning, barcode support for packaged foods, restaurant-specific menu integrations, multilingual support, offline allergen databases, user accounts with saved preferences, and nutrition analysis to make MenuSafe a complete dining companion.
Log in or sign up for Devpost to join the conversation.