# Patience

- **Event:** [RAISE Summit Hackathon](https://cerebralvalley.ai/e/raise-summit-hackathon)
- **When:** Jul 4 at 9:00 AM – Jul 5 at 7:00 PM (GMT+2)
- **Where:** Paris, France
- **Team:** [Patience Dusengimana](https://cerebralvalley.ai/u/DPatience11)
- **GitHub:** https://github.com/dpatience/LivestokOS/
- **Demo video:** https://www.youtube.com/watch?v=e73bYqtsb6k
- **Gallery:** https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/119

LivestokOS — Real-Time Situational Intelligence for Livestock Operations

Livestock farms are physical, distributed, high-stakes environments where risk builds faster than any person can track it: herds bunching at gates under heat stress, paddocks overgrazed past recovery, carbon and methane impact accumulating invisibly, reproductive windows missed. Farmers currently work from periodic checks and static records, not a live picture of what is happening across the land right now.
LivestokOS builds a live situational model of a farm from streaming inputs — GPS/collar telemetry, geofence crossings, satellite NDVI imagery, and rotation events — and uses that model to drive one plain-language, one-tap advisory a farmer can accept or override in the field, with every override feeding directly back into the next recommendation. Example: the system detects herd bunching near a gate under heat stress before it becomes a crush risk, and surfaces one clear next action instead of raw data to interpret. Advisory phrasing is generated on Crusoe Managed Inference; the underlying risk detection itself is a deterministic model built from live telemetry, not a model guessing from text.

Beyond that single moment, LivestokOS is the operating system underneath it: paddock geofencing and rotation-compliance scoring, satellite-informed grazing-recovery projections, a tamper-evident carbon and methane ledger that turns verified rotation and feed data into an auditable sustainability record, and a reproductive and lactation tracking module for the herd's female animals. The system is farm-mode aware, supporting both open-pasture and indoor zero-grazing operations differently rather than treating every farm the same way. Two installable, offline-first apps sit on top: one for the farmer in the field, with NFC-tag device pairing and quick-add daily logging, and one for administrators overseeing multiple farms' telemetry and equipment health.

Architecturally, the backend is an Elixir/OTP umbrella application: ingestion, digital twins, geofencing, satellite processing, and inference each run as independently supervised processes, so a slowdown or failure in any single subsystem never takes down the rest of the farm's operations. This is a deliberate fault-isolation design chosen for environments where connectivity and hardware are unreliable, not an afterthought bolted on later.
The goal is to give a non-technical farmer a trustworthy, continuously updating picture of a physical environment that changes by the minute, and exactly one decision they can act on immediately — with the system getting measurably sharper every time they do.

## More from RAISE Summit Hackathon

- [T2MV](https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/116)
- [MAYDAY](https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/117)
- [TOSCO](https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/118)
- [Fella](https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/120)
- [ATIPE](https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/121)
- [Call of Duty](https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/122)

---

Markdown version of https://cerebralvalley.ai/e/raise-summit-hackathon/hackathon/gallery/119. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt