# Nova

- **Event:** [WeaveHacks 3: Self-Improving Agents Hackathon with Weights & Biases](https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80)
- **When:** Jan 31 at 9:00 AM – Feb 1 at 5:00 PM (PST)
- **Where:** 400 California St, San Francisco, CA
- **Team:** [Eric M](https://cerebralvalley.ai/u/Eric_m_cv), [Shoe Shoemaker](https://cerebralvalley.ai/u/shoe), [Matt Ruiters](https://cerebralvalley.ai/u/ruiters)
- **GitHub:** https://github.com/mcr5fh/Nova
- **Demo video:** https://youtu.be/7zwUR-NkgY4
- **Gallery:** https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery/43

We don't want to automate decision making. Since AI can write code, everyone knows planning becomes the most important step. However, no one wants to spend the time to challenge their assumptions or think about edge cases. This results in additional code iterations that slows down progress.

We built a tool that does two things:
1. Implements a robust planning & implementation process before writing a line of code
2. Improves our planning & implementation automatically after a coding session is complete.

The goal is the reduce the amount of steering during building, and this comes from good planning, not better coding models.

---

Markdown version of https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery/43. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt