AGENT SYSTEMS
Perception
Vision Models
Detect, track and structure the visual signals that matter across people, objects, movement and events.
- Detection
- Tracking
- Evaluation
Deqa combines computer vision, VLM reasoning and an AI Agent Harness to turn existing video into governed action — across retail, warehouses, factories, infrastructure and field operations.
A governed AI Agent for the physical world.
Not a camera vendor. Not a one-off AI project. Deqa is building the agent infrastructure that lets machines perceive, reason and execute customer SOPs across the physical economy.
Saudi-built models and product IP
02Edge, on-premise, cloud or hybrid
03Agent Harness, customer SOPs and traceable evidence
Deqa connects perception, multimodal reasoning, agent orchestration and customer SOPs in one governed loop — from frame to accountable action.
Deqa’s AI Agent Harness surrounds the VLM with the context, tools, permissions and observability it needs to operate inside a real business.
Perception
Detect, track and structure the visual signals that matter across people, objects, movement and events.
Multimodal context
Understands scenes over time by combining video, language, site context and the customer’s operating question.
Governed orchestration
Connects the VLM to tools, memory, policies and enterprise APIs so reasoning becomes safe, observable action.
Executable operations
Turns customer SOPs into triggers, permissions, escalation paths and the evidence required to close every loop.
Deqa Box is the secure edge runtime for visual models, VLM reasoning and the local side of the Agent Harness. It keeps critical workflows fast and resilient while respecting privacy and bandwidth constraints.
The operational question changes. The intelligence layer compounds.
Journeys · queues · dwell · conversion
Flow · loading · dwell · evidence
Safety · quality · assets · SOPs
Attendance · tasks · roads · sites
The Agent Harness routes each visual event through the right context, SOP, permission and evidence path — without replacing your camera estate.
Existing cameras + video
Vision models structure the scene
VLM + AI Agent Harness
SOP action + traceable evidence
Live deployments, daily operating volume and product IP — stated clearly, without turning pilots or ecosystem relationships into customer claims.
Accelerated by national programs and relied on by facilities across the Kingdom.
Start with a measurable operating question, representative cameras and a deployment boundary that fits your environment.
A reusable VLM agent stack, edge hardware and SOP integrations — with recurring software and expansion by site, camera and workflow.
Deqa AI is a Saudi computer-vision company based in Riyadh. It turns the video from cameras a business already has into answers and actions: who is on site, how customers move, what a screen audience looks like, and whether a procedure was followed. It combines vision models, VLM reasoning, an AI Agent Harness and the customer’s own procedures (SOPs), and can run on the Deqa Box edge device, on-premise, in the cloud or hybrid.
Three products on one platform: Deqa Performance (workforce attendance with face verification, field teams and productivity), Deqa Experience (heatmaps, customer journeys, dwell time and sentiment in stores, restaurants and hotels) and Deqa Ads (digital screens that measure their real audience and adapt content). Deqa Box is the edge device that runs the AI next to the cameras.
Usually not. Deqa connects to the CCTV and IP cameras you already have, and to your existing HR and point-of-sale systems. A pilot starts with a few representative cameras.
You choose the deployment: at the edge on a Deqa Box next to the cameras, on-premise inside your network, in the cloud, or hybrid. Deqa is built for Saudi PDPL requirements; facial photos are not stored, only a secure biometric template for attendance.
Week 0: agree the operating question, baseline and site survey. Weeks 1–2: integration and calibration. Weeks 3–5: live measurement. Then you decide to expand, adapt or stop.
Plans start at 499 SAR per month for Deqa Performance (up to 25 employees), 799 SAR for Deqa Experience (one site) and 599 SAR for Deqa Ads (up to 5 screens). Annual billing gives two months free, prices exclude VAT, and enterprise plans are priced per project.
Yes. The platform, dashboards, reports and support are available in Arabic and English, and the product is designed for Saudi and GCC operations.