🗓️
2024
Cohalal
CoHalal is a lightweight food-scanning service designed for Muslim travelers navigating unfamiliar dining environments. Instead of relying on word-of-mouth or scattered reviews, the system analyzes menu items and highlights potential non-halal ingredients, reducing uncertainty and helping users make confident choices.
CoHalal started with a real user problem: a Saudi traveler repeatedly struggled to find halal-friendly restaurants during a long stay in Korea. Existing solutions relied on community posts, inconsistent reviews, or manual ingredient checking—none of which worked well for someone unfamiliar with the local language or food culture.
The goal was simple: allow users to scan a menu and instantly understand which items might not be halal. The system combines OCR-based text extraction with ingredient classification to detect non-halal components such as pork, alcohol, or ambiguous additives. Instead of offering a long nutritional breakdown, it focuses on clarity—green for safe, yellow for unclear, red for high-risk.
As a Builder, the project emphasized user-driven problem definition. Interviews with Muslim travelers shaped the core features: quick scanning, minimal text, and culturally clear signals. The MVP was tested as part of a tourism-venture competition, securing a top-200 placement and validating the concept’s relevance for inbound travel.
CoHalal reflects a broader belief that meaningful products come from understanding context deeply—language, culture, and small frictions that significantly impact real experiences. It remains an exploration of how AI can support accessibility across cultural and linguistic boundaries.
