# Good Omen

- **Event:** [Gemini 3 Hackathon London](https://cerebralvalley.ai/e/gemini-3-hackathon-london)
- **When:** Sat, Dec 13 at 9:00 AM – 10:00 PM (UTC)
- **Where:** CodeNode, London, UK
- **Team:** [aaron jones](https://cerebralvalley.ai/u/aivideoguy)
- **GitHub:** https://github.com/yepicaiaaron/Gemini3-Hackathon-
- **Demo video:** https://link.yepic.ai/liz0
- **Gallery:** https://cerebralvalley.ai/e/gemini-3-hackathon-london/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/gemini-3-hackathon-london/hackathon/gallery/33

ATLAS is an open-source, long-running, multimodal video production system that automatically turns complex ideas, research, and data into fully produced explainer videos.

ATLAS works as a marathon agent: it researches sources, plans multi-scene narratives, generates and edits high-resolution visuals, produces lifelike avatar videos on self-hosted GPUs, applies motion design and effects, and assembles everything into a final video. Each output is composed of multiple 2–10 second scenes, allowing ATLAS to maintain visual, character, and narrative consistency across the entire production.

Unlike one-shot video generators, ATLAS is built for multi-step, compute-heavy workflows where state, quality, and coherence matter.

ATLAS is not a video tool. It’s a production system.

---

Markdown version of https://cerebralvalley.ai/e/gemini-3-hackathon-london/hackathon/gallery/33. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt