Loading...
© 2026 All rights reserved.
A cross-platform mobile application that scans food products using the device camera and analyzes ingredients through AI. Provides neutral, science-based nutritional information without brand bias or alarmist health scoring, compliant with Tunisian consumer protection regulations.
mobile
Open Source
6 tools
Featured Project
This project was created to address the need for a neutral, science-based food analysis tool that respects Tunisian consumer protection laws and serves Arabic-speaking users who are underserved by existing nutrition apps.
Consumers struggle to interpret food labels and make informed dietary choices. Existing apps often use alarmist language, brand-biased scoring, or lack support for Arabic languages and Tunisian regulatory context.
Sole developer — designed the app architecture, implemented the Flutter frontend with Riverpod state management, integrated Google Gemini AI for ingredient analysis, built the Supabase backend with PostgreSQL, and managed cross-platform deployment.
Developed a cross-platform mobile app using Flutter that leverages Google Gemini 2.5 Flash for ingredient analysis, providing neutral and scientific information. The app supports four languages with RTL layout, operates in guest mode without account requirements, and offers personalized health profiles for dietary condition management.
OCR accuracy varied significantly depending on label quality, lighting conditions, and font types. The Gemini API occasionally returned inconsistent formatting for parsed ingredients, requiring additional sanitization logic.
A fully functional cross-platform food scanner with AI ingredient analysis, multilingual support, guest mode, and health profile integration deployed as a PWA on web and available on Android and iOS.
Gained deep experience with Flutter state management and cross-platform deployment. Learned to handle edge cases in OCR preprocessing and API response validation.
Scanni allows users to scan food product labels using their device camera and receive a detailed AI-powered analysis of ingredients, E-codes, and nutritional composition. The app uses Google Gemini 2.5 Flash for ingredient recognition and analysis, presenting results in a clean, neutral format. It features a full guest mode for scanning without account creation, secure cloud sync via Supabase for registered users, and personalized health profiles where users can track ingredients related to specific conditions (celiac, diabetes, hypertension). The app supports four languages (English, French, Arabic, Tunisian Arabic) with full RTL layout support for Arabic variants and is designed as a Progressive Web Application for broad accessibility.
A data analysis and machine learning project exploring the World Happiness Report dataset. Uses linear regression, polynomial regression, and K-Means clustering to identify factors influencing global happiness scores and group countries by well-being patterns.
A modern web-based room reservation and management system built with FastAPI and React. Allows organizations to manage bookable spaces, browse available rooms, make reservations with interactive calendar views, and provides admin tools for managing rooms, equipment, users, and reservations.