# Tamachi

- **Event:** [Gemini 4 NYC Hackathon](https://cerebralvalley.ai/e/gemini-4-nyc)
- **When:** Sat, Nov 21 at 9:00 AM – 10:00 PM (EST)
- **Where:** New York, NY
- **Team:** [Andrew Park](https://cerebralvalley.ai/u/aardpark)
- **GitHub:** https://github.com/aardpark/tamachi
- **Demo video:** https://youtu.be/wEdM5PuXgDA
- **Gallery:** https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/70

Tamachi gently realigns agents when they're working with stale context within a repo. 

Agentic race conditions are the bane of multi agent workflows, this aims to cut down on issues where multiple agents working on a repo hurts more than it helps. The goal is to allow for truly large and sophisticated swarms of agents to improve productivity and throughput while also keeping costs down and making iteration and development faster.

## More from Gemini 4 NYC Hackathon

- [Geminate](https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/67)
- [DropShield](https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/68)
- [Aether](https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/69)
- [AlwaysLateToEverything](https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/71)
- [Shadow Market](https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/72)
- [foresight-worlds](https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/73)

---

Markdown version of https://cerebralvalley.ai/e/gemini-4-nyc/hackathon/gallery/70. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt
