# Arroztek

- **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:** [Brayan Roberto Ccarita Cruz](https://cerebralvalley.ai/u/arrozdev)
- **GitHub:** https://github.com/hallzyx/cut-sense
- **Demo video:** https://youtu.be/3dt39OKoV4U
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1427

CutSense turns a photo of an irregular flat offcut (leather, felt, foam, cardboard) into a cutting plan that has been checked by an independent geometric verifier, and then helps reuse what is left.

The problem: makers and small workshops keep drawers of offcuts that are too good to throw away but too awkward to plan around. The loss is measurable: a 2024 study of garment manufacturing estimated that 16–21% of fabric was wasted in the cut-and-sew process (Abou-Chakra et al., 2024). Fitting pieces onto an irregular shape with flaws by eye is slow, and it is hard to know whether a layout really fits or what can still be made from the remainder.

How it works: AI outlines the scrap from the photo and you correct it and mark the defects. You calibrate the scale and pick what to make, from ready-made kits to your own shapes or a design worked out with Hilo, an AI pattern-maker chat that also shows an approximate 3D assembly preview. A deterministic nesting engine places the pieces and an independent verifier checks that nothing overlaps, leaves the material or touches a defect before you see the plan. Then you export it as SVG, PNG, a 1:1 print across paper sheets or a DXF.

The leftover loop: CutSense prefers layouts that leave the largest single free fragment, lets you plan another product on it, or start again from a photo of the real remnant. AI proposes, deterministic code decides and verifies, and every result is reproducible. It is open source (MIT) and ships as a single Docker container.
---

Reference:
Abou-Chakra, K., Archipov, K., Berkovitz, S., Perry, E., & Spellenberg, R. (2024). Analysis of pre-consumer textile waste within the apparel supply chain. Bren School of Environmental Science & Management, University of California, Santa Barbara. https://bren.ucsb.edu/projects/analysis-pre-consumer-textile-waste-within-apparel-supply-chain

## More from Meta Global AI Developer Hackathon

- [Rastro](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1424)
- [Eternity Nest](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1425)
- [Abhi-the-great](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1426)
- [Competition Ops Agent](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1428)
- [Creative Platform](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1429)
- [honeycomb.](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1430)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/1427. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt