# Musory

- **Event:** [Agents in Context Toronto: a MongoDB Hackathon](https://cerebralvalley.ai/e/mongodb-toronto-hackathon)
- **When:** Sat, Nov 14 at 9:00 AM – 9:00 PM (EST)
- **Where:** Toronto, Canada
- **Awards:** 3rd Place
- **Team:** [Kaung Zin Hein](https://cerebralvalley.ai/u/zin_kg)
- **GitHub:** https://github.com/Zen-cronic/musory
- **Demo video:** https://youtu.be/Q6TSA8ha3UM
- **Gallery:** https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/6

Venues play one playlist for everyone, and the apps that let guests shape the music keep their likes on view with no easy way to take them back. Musory: your venue learns what its crowd likes and makes music for it, and one tap makes every venue forget you.

I love electronic music. Not everyone does. I go clubbing, and I don't think I will pull out Shazam mid-rave.

Scan the venue's QR code, choose whether your likes count here, then swipe right to like or left to skip, or type what you want to hear. A venue only sees likes pooled from 3 or more people, never names. When the crowd's taste shifts, ElevenLabs Music makes a new track and the room fades into it.

In Rockbot, a bar music app, others may see your song votes and check-ins by default, and deleting them takes a request. In Musory, your phone shows the next track each room would play without you. Tap forget me, and one step deletes your likes, your taste and any track made from them, at every venue. About 2 seconds later, each room chooses its music without your taste.

Toronto nightlife is shrinking (CBC), so venues need guests to come back. This is a prototype with simulated venues: Copper Bar and Pulse Club are two screens on one laptop, and the regulars are test accounts. A venue pilot would test whether guests prefer its choices. Built November 14. The demo video is narrated with ElevenLabs text to speech in a clone of my own voice.

The three questions, answered. Leaking: each venue has its own database login that reads only its own View, and Pulse Club reading Copper Bar's crowd memory is refused by Atlas (error 13). Wrong or expired: skip it, set old likes aside, or count likes for tonight only (Atlas TTL), and forget me deletes your likes, taste and tracks made from them in one transaction. Proof: every pick is saved next to the pick the same search makes with no memory. In our scripted test, memory changed the pick in 8 of 9 scenarios, and a guest who didn't consent changed nothing.

## More from Agents in Context Toronto: a MongoDB Hackathon

- [One Lesson](https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/3)
- [Solo Spades](https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/4)
- [Sharleen Chebet](https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/5)
- [Phoenix](https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/7)
- [MongoDB Larpers](https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/8)
- [Sepi](https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/9)

---

Markdown version of https://cerebralvalley.ai/e/mongodb-toronto-hackathon/hackathon/gallery/6. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt