# fix-it-in-post

- **Event:** [fal x Sequoia 72-Hour Video Hackathon](https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon)
- **When:** Jul 17 at 12:00 PM – Jul 19 at 8:00 PM (PDT)
- **Where:** Online
- **Team:** [Erik E](https://cerebralvalley.ai/u/erike)
- **GitHub:** https://github.com/eerwitt/fix-it-in-post
- **Demo video:** https://www.youtube.com/watch?v=f4P46PxIXNA
- **Gallery:** https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon/hackathon/gallery/48

What if your AI video had an editor watching every frame? GenAI can produce an amazing sequence while hiding a few moments where the motion breaks, or a rabbit suddenly appears in the background. This project connects V-JEPA 2.1 to my fal workflow as an automated rough-cut editor, comparing generated motion with JEPA predicted animation and flagging the exact moments where things may have gone off the rails. Instead of manually searching through every frame, creators and AI agents get a map of where in the frame to look, what to review, and which sections may need regeneration. The result is a reusable validation layer for generative video: not another model making content, but one improving the quality of the generated content.

---

Markdown version of https://cerebralvalley.ai/e/fal-x-sequoia-virtual-hackathon/hackathon/gallery/48. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt