The ecosystem
AIREZ is not defined by how much it senses. It is defined by what it understands.
80+ sensor modalities across 12 intelligence categories are only the first layer. The ecosystem is a five-layer model that carries an observation all the way to a response — and learns from the outcome.
Every layer of this ecosystem converges into one intelligence. Speak with it directly and ask how AIREZ would understand your environment.
Begin the Conversation01
Awareness
Everything that observes the physical environment.
02
Intelligence
Multiple AI models orchestrated per task.
03
Context
Signals fused into a single understanding.
04
Action
Autonomous, system and human response.
05
Continuous Learning
Every deployment expands the platform.
Layer 01 — Awareness
Everything that observes the physical environment.
Grouped into intelligence categories rather than one long specification list. Each category expands as customer environments require it.
80+
Sensor modalities across 12 categories
Visual Intelligence
AvailableOptical and beyond-visible imaging that gives every other signal a scene to sit in.
- Camera (RGB / IP)Visual scenes for object, incident and operational detection.
- Thermal cameraHeat signatures for people, vehicles, fire and overheating equipment.
- Low-light / night visionImaging in extremely low light environments.
- Multispectral cameraMultiple wavelengths for environmental and industrial analysis.
- Hyperspectral cameraChemical signatures, vegetation stress and material composition.
- Optical image sensorLight converted to electronic signal for machine vision.
- LiDARLaser pulses building precise 3D maps of space and objects.
- Laser rangefinderDistance and object positioning.
- Optical beam (photoelectric)Beam-break detection, often on perimeters.
- Light / illuminationAmbient light for triggers and camera optimization.
- Near infrared780–1700 nm imaging and material analysis beyond visible light.
11 modalities supported in this category today. Categories expand as customer environments require them.
Layer 02 — Intelligence
AIREZ selects the right intelligence for each task.
Model providers are supporting technologies. Routing, context, governance and provenance stay with AIREZ.
Supporting AI platforms
AIREZ Intelligence Layer
Orchestration selects the right capability for each task.
Model providers are supporting technologies. Routing, context, governance and provenance stay with AIREZ, so providers can be added or retired without changing a deployment.
Right model, right task
Every request is routed by task class, latency budget, sensitivity and cost — not by vendor preference.
Model-independent by design
Providers can be added, swapped or retired without changing deployments, integrations or workflows.
Edge and cloud together
Lightweight perception runs at the edge; heavier reasoning escalates only when it earns its place.
Human-in-the-loop
Autonomy is bounded and policy-governed. Confidence and provenance travel with every conclusion.
Layer 03 — Context
Sensor fusion is where monitoring becomes understanding.
No single signal is conclusive. Combined, they describe a situation — with confidence and provenance attached.
Why it matters
Speed and trajectory read against surface conditions and visibility, so a near-miss is understood as risk rather than logged as motion.
No single sensor in this pattern is conclusive on its own. Context is the product.
Layer 04 — Action
The operational outcome of awareness, intelligence and context.
Autonomy is bounded and policy-governed. Every response path can be reviewed, escalated or overridden by a person.
Robotics
AutonomousDispatch a platform to verify before a person is committed.
Quadrupeds
AutonomousTerrain-capable inspection where fixed sensing cannot reach.
Drones
AutonomousAerial confirmation of a wide-area or perimeter event.
Humanoids
AutonomousEmerging response class for human-shaped environments.
Voice
HumanConversational briefing and command through the AIREZ assistant.
Notifications
HumanThe right person, with the context already assembled.
Building automation
SystemHVAC, lighting and airflow driven by understanding.
Access control
SystemDoors, gates and credentials adjusted to the situation.
Mobile applications
HumanField teams carrying the same picture as the GSOC.
Smart glasses
HumanHands-free context delivered where the work happens.
Human escalation
HumanBounded autonomy always ends with a person in authority.
Layer 05 — Continuous Learning
The ecosystem is designed to stay unfinished.
Every deployment expands what the platform knows. Growth is additive — never a migration, never a redesign.
Every deployment teaches the platform
New environments expose new patterns. What is learned in one site raises the baseline for the next.
Every sensor class is additive
New modalities join the awareness layer as connectors — never as a migration or a redesign.
Every model can be replaced
Orchestration means better intelligence can be adopted without changing deployments or workflows.
Every outcome refines context
Confirmed and dismissed conclusions tune fusion, thresholds and the confidence attached to future ones.
The complete architecture
The Physical Intelligence Layer, end to end.
From the physical world to a response — and back again as learning.
- 01Physical WorldSites, assets, people and conditions
- 02Awareness70+ sensor types across 12 categories
- 03Edge IntelligenceLocal perception where latency matters
- 04AIREZ PlatformNormalization, routing, governance
- 05Multi-Model AIOrchestrated reasoning per task
- 06Sensor FusionSignals resolved into one situation
- 07GSOC IntelligenceOperational picture and priority
- 08ActionAutonomous, system and human response
- 09Continuous LearningEvery outcome improves the layer
Need a sensing class we don't list yet?
Most integrations begin as a customer request.