# Zān

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [Aishwarya Murugappan](https://cerebralvalley.ai/u/gracyashhh)
- **GitHub:** https://github.com/gracyashhh/zan-kitchen-companion
- **Demo video:** https://youtu.be/ZgeEi1n6tYU
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4895

Zān is an AI-powered kitchen companion designed to make everyday food decisions simpler, reduce unnecessary grocery purchases and help people make better use of ingredients they already own.

The problem is familiar: we forget what's in our fridge, struggle to decide what to cook, buy ingredients we already have and find food tracking tedious.

Zān connects five everyday kitchen activities in one application.

My Fridge: Users photograph their refrigerator. Meta's Muse Spark identifies visible ingredients and estimates quantities. Users can review uncertain detections, correct mistakes and save the results to their personal inventory.

Smart Recipes: Users select available ingredients and receive AI-generated recipes with ingredients, instructions and missing-item suggestions. Recipes can be saved, revisited and read aloud.

Shopping List: Users can add missing recipe ingredients directly or speak multiple grocery items in one recording. Meta's speech transcription and language understanding help convert natural speech into individual shopping items.

Food Diary: Users photograph meals, review detected foods and portion estimates, and receive approximate nutritional information using USDA food reference data. Previous meals and photos remain accessible.

Personal Experience: Supabase authentication, private storage and persistent data keep each user's kitchen organised across sessions. The responsive Streamlit interface supports both desktop and mobile use.

A key design principle is human-in-the-loop AI. Rather than assuming every model prediction is correct, Zān allows users to review and correct AI outputs before saving them.

Zān is more than a recipe chatbot. It creates a connected workflow from identifying food to deciding what to cook, preparing a shopping list and logging meals.

The goal is simple: less food waste, less mental effort and a more mindful kitchen.

Live Demo: https://zanfan.streamlit.app/

Zān is fully deployed and publicly accessible through Streamlit Community Cloud. Judges can explore the working application, including AI-powered fridge scanning, personalised recipes, voice-enabled shopping lists and food diary tracking.

## More from Meta Global AI Developer Hackathon

- [Argus Labs](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4892)
- [Open Intelligence](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4893)
- [MIDUSAFE Sentinel](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4894)
- [Mathew Ukoh](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4896)
- [HearthAI](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4897)
- [Ethan Wilson](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4898)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4895. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt