Food discovery was treated as a core problem, not a secondary feature. Many users don’t struggle to order food because of lack of options, but because they don’t know what they want in the moment. Decision fatigue, mood, and health considerations often make browsing menus feel overwhelming rather than helpful.

To address this, we explored an AI-assisted discovery experience that helps users decide what to eat based on context rather than endless scrolling. The system combines available user health data (when provided), such as dietary preferences or restrictions, with real-time emotional context. Users could describe how they’re feeling in natural language, or optionally allow the front camera to capture facial cues to infer emotional state. These signals are used together to generate food recommendations that feel relevant to the user’s current mindset.

Instead of presenting static lists, the experience reframes discovery as a guided suggestion process. Recommendations adapt based on mood, appetite, and personal constraints, helping users move from uncertainty to choice more quickly. This approach shifts food discovery from search-heavy behavior to a more intuitive, supportive interaction—making it easier for users to decide what to eat and increasing confidence in their selection.