AI Agricultural Intelligence Platform
Transforming agricultural and environmental data into intelligent decisions using Artificial Intelligence, Computer Vision, Digital Twins, GIS, and autonomous systems.
The intelligence platform is the product. Hardware is one future data source.
The Gulf feeds itself against every constraint.
Climate, pests, and water scarcity converge faster than traditional monitoring can respond.
of GCC food is currently imported
This is the strategic vulnerability the National Food Security Strategy 2051 targets.
The red palm weevil arrived in the region
Larvae feed inside the trunk before any surface symptom is visible.
Every liter of irrigation carries a desalination cost
Per tree precision turns water from a ceiling into a lever.
An AI that flies.
FeraSky is not a camera drone that sends data to a server. The Jetson Orin Nano Super runs the full detection model in flight, so diagnosis happens above the tree at the moment of capture. Every generic agricultural drone flies a route and drops raw imagery for later review. FeraSky delivers the finished diagnosis while still in the air.
Sees
RGB, NIR, and UV inspection light.
Thinks
A trained detection model runs on the Jetson Orin Nano Super at 67 TOPS, in flight, with no cloud.
Decides
Every canopy receives a labeled diagnosis, a confidence score, and a treatment recommendation, geotagged and streamed live.
Why FeraSky is different
Four design choices that separate an intelligence platform from a camera on wings.
No cloud round trip
Zero second latency because the model runs onboard.
Per tree precision
Precision applied per tree, not per field.
Sovereign deployment ready
A Jetson local fallback keeps data in country when the Ministry requires it.
Crop agnostic
Date palm, citrus, mango, olive, stone fruit, apple, pomegranate, avocado, and banana.
A unified intelligence stack
Nine interconnected systems, engineered to work as one platform rather than nine separate tools.
AI Platform
The core reasoning layer that orchestrates models, data, and decisions across the system.
Computer Vision
Detection, segmentation, and classification models trained on agricultural and environmental imagery.
AI Agent
Autonomous reasoning agents that plan, monitor, and recommend without constant human input.
Digital Twin
A continuously updated virtual replica of fields, terrain, and infrastructure.
GIS Engine
Geospatial infrastructure that anchors every insight to real world coordinates.
Mission Planning
Automated planning for data collection missions across land and airspace.
Analytics
Dashboards and reporting that translate raw signals into operational metrics.
Telemetry
Live data streams from sensors and future autonomous assets in the field.
Future Robotics
An extensible foundation for ground and aerial robotics as hardware integration matures.
From raw data to real decisions
A six stage intelligence cycle designed to run continuously across a growing season.
Collect
Ingest imagery, sensor readings, and geospatial data from multiple sources.
Analyze
Apply computer vision and AI models to detect patterns and anomalies.
Understand
Fuse findings into the digital twin for full contextual understanding.
Predict
Forecast trends in crop health, water stress, and environmental risk.
Recommend
Generate clear, prioritized recommendations for the people on the ground.
Act
Deliver intelligence into workflows where decisions are made and executed.
Intelligence across the entire landscape
A single platform designed to extend across crops, terrain, water, and infrastructure.
Crop Monitoring
Continuous visibility into crop health and growth stages.
Disease Detection
Early identification of disease signatures in plant canopy.
Pest Detection
Pattern recognition for early pest pressure indicators.
Water Stress
Identification of irrigation gaps before yield is affected.
Tree Counting
Automated counting and health scoring across orchards.
Yield Analytics
Data driven modeling to support yield estimation.
Environmental Monitoring
Tracking environmental indicators across a monitored region.
Infrastructure Inspection
Visual inspection support for irrigation and rural assets.
Land Mapping
High resolution mapping of terrain and land use.
Decision Support
Prioritized guidance generated from fused intelligence.
Geospatial Analytics
Spatial analysis layered across every field and region.
Simulation
Digital twin simulation used to test scenarios before deployment.
How FeraSky earns
Three revenue streams, each designed around a different stage of the platform.
SaaS dashboard
Per hectare, per monthThe primary recurring revenue stream, a live intelligence layer delivered as software.
Hardware sale
Per drone unit, from the hardware rollout stage onwardAn extension of the software, not the primary product.
Service contracts
Per mission and per seasonStructured engagements for government and enterprise pilots.
How FeraSky stands apart
Positioned against the agricultural drone platforms most commonly deployed in the region today.
| Onboard AI in flight | RGB, NIR, and UV | Sovereign deployment | Crop agnostic | Live geotagged uplink | Software first platform | |
|---|---|---|---|---|---|---|
| FeraSky | Included | Included | Included | Included | Included | Included |
| DJI Agras | Not applicable | Not applicable | Not applicable | Not applicable | Not applicable | Not applicable |
| XAG | Not applicable | Not applicable | Not applicable | Not applicable | Not applicable | Not applicable |
| JOUAV | Not applicable | Not applicable | Not applicable | Not applicable | Not applicable | Not applicable |
The addressable market
Sizing that has been designed, not claimed.
GCC precision agriculture
The addressable market across the UAE, Saudi Arabia, Qatar, Bahrain, Kuwait, and Oman.
UAE, Saudi Arabia, and Qatar orchards
Orchard hectares reachable within the pilot geography.
GCC date palm and citrus orchards
The first pilot geography, determined jointly with the host country's ministry of agriculture.
Sizing methodology available on request. No revenue is claimed.
Built first in software, validated by simulation
FeraSky is engineered as a software first platform, with hardware treated as an extension of that intelligence.
Software Platform
Substantially completeThe core AI, computer vision, GIS, and digital twin platform has been designed and developed.
Simulation Environment
CompletedA simulation first approach was used to develop and test the platform before hardware integration.
Detection Pipeline
OperationalThe crop disease detection pipeline runs from image capture through to a labeled diagnosis.
Hardware Integration
In preparationThe platform is being prepared to integrate with drones and future robotic hardware.
Field Validation
Awaiting fundingFunding will accelerate the transition from simulation to real world field validation.
All figures on this page describe current development status. No performance results, pilots, or commercial deployments are claimed until validated in the field.
What is built and verifiable today.
A working inventory of the platform as it stands, not a forecast.
Detection pipeline operational
The crop disease detection pipeline runs end to end today.
Live testing panel available
Available at ferasky.com/dashboard/demo.
No customers, revenue, or paid pilots are claimed. All items marked complete are verifiable in the public repository.
The stack powering the platform
A modern, production grade technology foundation chosen for rigor and scale.
Python
The core language across the AI and platform codebase.
Named systems, not generic claims
- Vision model
- MobileNetV2 fine tuned on PlantVillage, 38 classes
- Agronomist LLM
- Llama 3.1 8B Instruct served through NVIDIA NIM, with a Jetson local Gemma 2B fallback
- Onboard compute
- NVIDIA Jetson Orin Nano Super, 67 TOPS
- Twin spectrum vision
- RGB and NIR, producing NDVI
- UV inspection light
- Canopy fluorescence, detecting fungal signatures before RGB symptoms appear
- Live uplink
- 4G LTE and MQTT, geotagged
- Autonomous flight
- PX4 with GPS waypoints
The path from software to a global platform
A deliberate, staged path from a complete software foundation to global deployment.
Software Platform
Core AI, vision, GIS, and digital twin platform designed and built.
CompleteFunding
Securing grants, investment, and strategic partners to accelerate the roadmap.
CurrentHardware and Flight Validation
Integrating drones and sensing hardware, then moving validation from simulation into controlled real world flight.
UpcomingPilot Projects
Structured pilots designed with partners and research collaborators.
UpcomingCommercial and Regional Deployment
Preparing the platform for its first commercial deployments, then extending validated deployments across additional regions.
UpcomingOne intelligence layer across every source
The long term architecture connects every signal, from orbit to the ground, into a single decision layer.
Hardware is one input among many. The platform is designed so that new data sources extend the same intelligence core rather than replace it.
The team
Mohamad Omar
Founder of USTech AI
A US educated engineer, University of Minnesota.
A decade across artificial intelligence, computer vision, autonomous drone systems, and national scale network infrastructure.
Founder of USTech AI, building sovereign AI platforms for the region.
Committed to relocating long term and building the team out of the GCC.
Intelligence is the product
Mission
To give every farmer, every ministry, and every food security council an intelligent layer that turns agricultural data into decisions.
Vision
The operating system for intelligent agriculture across the Gulf and beyond.
Core Philosophy
The intelligence platform is the product. Hardware is one future data source, an extension of the software, never the other way around.
FeraSky is an intelligence platform. Drones and future robotics are extensions of that platform, not the platform itself.
FeraSky is bootstrapped to date by USTech AI. No external capital has been raised.
Building FeraSky across the region
The commitments that anchor this platform to the region, not just to a market.
Regional Presence
Founder committed to building the team out of the GCC, with the specific host city determined by the granting authority.
Pilot Orchards
Initial deployments across GCC orchards, in coordination with the ministry or authority of the host country.
Ecosystem Alignment
Direct engagement with national food security councils, investment offices, and innovation funds across the UAE, Saudi Arabia, and Qatar.
National Contribution
FeraSky provides the ground truth geospatial data layer that a national food security dashboard can be built from, aligned with UAE Strategy 2051, Saudi Vision 2030, Qatar National Vision 2030, and UAE Net Zero 2050.
Let us build the future of agricultural intelligence together
We welcome conversations with government programs, investors, accelerators, and research partners.