# Project export: Snorkel

This document was generated by HackStack to give an AI agent context about a hackathon project. Sections are labeled with their provenance; content marked as truncated was cut to keep this document small.

## Project metadata

- Hackathon: UC Berkeley AI Hackathon 2026
- Tagline: Validate your ideas using AI simulation.
- Devpost: https://devpost.com/software/snorkel-upla5v
- GitHub: not linked
- Video: https://www.youtube.com/embed/aGpkrl6PYyk?enablejsapi=1&hl=en_US&rel=0&start=&version=3&wmode=transparent
- Team: contributor stats unavailable

## Devpost submission (written by the team)

### Inspiration

Inspired by the friction of gathering rapid customer feedback, Snorkel was built to simulate realistic user reactions and validate Product-Market Fit using the MiroFish swarm intelligence engine. To orchestrate this multi-agent environment, we developed a full-stack application featuring a fast Vue frontend and a Python Flask backend. We gave the AI personas persistent, long-term memory across simulation rounds by integrating an Agent Memory Server backed by Redis. With how expensive multi-agent systems can get, we leveraged The Token Company's compression middleware to save precious tokens and cost. We also integrated Browserbase's software so that our agents can use any web tools necessary throughout he simulation. Due to our complex backend and bunch of API routes proxying for agents, we use Sentry to observe these endpoints. The entire architecture is containerize with Docker to ensure easy deployment of the Python, Node.js, and Redis services. Challenges Building this system presented complex challenges, specifically around configuring the persistent memory database and managing the orchestration of multiple AI personas. We also had to carefully manage environment configurations to ensure smooth communication between the Vite proxy, the backend API, and the background workers. Getting everything setup like Redis locally was a pain especially, so a lot of time was spent debugging this. Conclusion Ultimately, developing Snorkel deepened our understanding of chaining LLMs to create realistic multi-agent interactions. It also improved our ability to jump into new codebases and start coding right away, as we integrated various sponsors.

## README (from the GitHub repository)

No README available.

## Detected evidence (automated analysis)

No repository was indexed for this project. Claimed technologies below could not be checked against code.
- Docker (technology) — claimed on Devpost, not found in the code
- Flask (technology) — claimed on Devpost, not found in the code
- JavaScript (language) — claimed on Devpost, not found in the code
- Node.js (technology) — claimed on Devpost, not found in the code
- Python (language) — claimed on Devpost, not found in the code
- Redis (technology) — claimed on Devpost, not found in the code

## Codebase structure

No repository index available.

## Key source files

No repository index available; no source files included.