# Croissants enjoyer

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [Cedric Martz](https://cerebralvalley.ai/u/eternal_baguette)
- **GitHub:** https://github.com/Cedric-Martz/EIDA
- **Demo video:** https://www.youtube.com/watch?v=m0Ydo3J3f24
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2394

Understanding how a physical device is built, and how to safely take it apart, is still surprisingly difficult. Technical documentation is scattered across repair guides, photographs, manuals, and CAD files. Even when this information exists, translating it into a clear understanding of a device's internal structure requires significant time and expertise. AIs can be struggleing to create 3D visuals of objects.

EIDA (Evidence-based Intelligence for Disassembly & Analysis) tries to bridges that gap.

EIDA is an AI-powered, local-first reconstruction workbench that transforms device photographs and verified Internet's sources into interactive, explorable 3D assemblies.

Users select a device, review automatically discovered technical documentation, and provide complementary photographs. EIDA then combines AI-powered component identification, visual segmentation, and geometry reconstruction to generate individual 3D parts that can be inspected in assembled, exploded, and workbench views.

What makes EIDA different is its evidence-based approach. Instead of presenting AI-generated geometry as truth, EIDA distinguishes visually supported information from inferred dimensions, hidden surfaces, and estimated component placement.

The goal is to make technical understanding more accessible, transparent, and interactive, helping repair enthusiasts, students, engineers, and researchers explore complex devices without needing to manually reconstruct every component.

EIDA turns scattered technical evidence into an interactive understanding of how things are built.

## More from Meta Global AI Developer Hackathon

- [ByteHive Supervisor](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2391)
- [void catalyst](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2392)
- [Arosa](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2393)
- [CanisMajor](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2395)
- [Muse RAISE](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2396)
- [RightOnTOp](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2397)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/2394. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt