machine viewenglishraw: /cases/animal-protein.mdbuild: d68074d2026-08-05T14:02Z
A company in the animal-protein industry is building a marketplace for meat sales. It can walk into an investor room with a working product: a slaughterhouse that ran the pilot, sellers who used it, buyers who responded.
200 Carcasses a day at the pilot facility · 1 Field-validated pilot · 2 Squads running in parallel
The first slaughterhouse partner processes 200 carcasses a day and sells to butcher shops, supermarkets, and restaurants across southern Brazil. The entire sales operation ran through a small team making calls, sending carcass photos over WhatsApp, and pushing weekly promotions by hand.
Carcass marketplace: Live storefront with photos, prices, and orders. No more back-and-forth
Controlled inventory model: Buyers see carcasses; the slaughterhouse keeps allocation control
AI quality-analysis system: OAK-1 devices analyze each carcass’s quality and feed the storefront
Internal operations layer: Stock, order flow, and allocation rules the internal team runs day to day
Related offering: [AI in the Operation](https://bleu.builders/offerings/ai-transformation/)
Contact: [Start a conversation](https://bleu.builders/contact/)
Animal protein · FIELD-VALIDATED PILOT
Animal protein
A company in the animal-protein industry is building a marketplace for meat sales. It can walk into an investor room with a working product: a slaughterhouse that ran the pilot, sellers who used it, buyers who responded.
200
Carcasses a day at the pilot facility
1
Field-validated pilot
2
Squads running in parallel
The problem
The first slaughterhouse partner processes 200 carcasses a day and sells to butcher shops, supermarkets, and restaurants across southern Brazil. The entire sales operation ran through a small team making calls, sending carcass photos over WhatsApp, and pushing weekly promotions by hand.
Our role
The request was a marketplace with AI quality analysis of carcasses. We went to the slaughterhouse first; then we built the minimum that answers the question. Bleu ran the field research, split the work into two decoupled squads, and took validation to real buyers while the founders were deep in fundraising.
What changed
The marketplace went live first, on team-uploaded photos, while the quality-analysis AI advanced in parallel. The result is a working pilot, validated with sellers and buyers in the field.
- Pilot live: a working slaughterhouse, its sellers, and its buyers
- Marketplace validated while the quality-analysis AI was still training
- Usability findings changed flows before the full build, not after
“Much of what was interviewed, we already knew from our own experience of practically 10 years in this market, but it hadn’t been put on paper. It brought very interesting insights. I am very satisfied with the entire process, with how you engaged and dove headfirst into this project.”
Client
Animal protein
FIELD-VALIDATED PILOT
What we've shipped
The marketplace launched without waiting for the quality-analysis model.
Carcass marketplace
Live storefront with photos, prices, and orders. No more back-and-forth
Controlled inventory model
Buyers see carcasses; the slaughterhouse keeps allocation control
AI quality-analysis system
OAK-1 devices analyze each carcass’s quality and feed the storefront
Internal operations layer
Stock, order flow, and allocation rules the internal team runs day to day
Areas we run
FIELD VALIDATION FIRST
Week one happened at the slaughterhouse, not in a planning doc.
Go to the operation
One week inside the facility watching the process happen: loading trucks, choosing carcasses, handling orders. Interviews with the internal team, sellers, and end buyers; route data, order history, and the full catalog collected before agreeing to a plan.
- Fieldwork before any build plan was agreed
- The storefront tested with butcher shops and supermarkets
- Flows changed before committing to the full build
TWO SQUADS, DECOUPLED
The marketplace could not wait for the AI to train.
Split and decouple
One squad on the AI system that analyzes carcass quality, one on the marketplace. The MVP launched on founder-uploaded photos so buyers could use the storefront while the model was still in training.
- Marketplace validation decoupled from AI training entirely
- OAK-1 devices and automatic quality analysis feeding the storefront
- Buyers browse carcasses without locking a specific one
RELATED OFFERING
See how Bleu structures this kind of work
The case shows what happened in one partnership. The offering explains the entry point, scope, and first delivery.
Bring us the problem.
Tell us what is going on. After you send it, the calendar opens so you can choose 15 minutes.
15 MINUTES · A CLEAR NEXT STEP.