Pilgrims served (1446H)
13.5Macross season
Crowd intelligence, pilgrim experience, and mass-event operations across the Haramain. Built for the Ministry of Hajj's Vision 2030 target — 30 million Umrah pilgrims annually — without compromising on safety, dignity, or the spiritual character of the rite.
Six pressures unique to mass-religious-tourism at the scale of the Haramain.
The Tawaf, Sa'i, and Jamarat see densities that outpace any concert, stadium, or transit hub on earth. Real-time density monitoring at one-meter granularity is non-negotiable; reactive crowd control is too late.
The Hajj rite compresses a year of preparation into 120 hours of uninterrupted operation. Every system has to be hardened to five-nines through that window — there is no maintenance pause.
Pilgrims arrive from 180+ countries. Wayfinding, alerts, and assistance must be delivered in fourteen-plus languages — Arabic, Urdu, Indonesian, Turkish, Bahasa, French, Hausa, and more.
August Hajj seasons see ambient temperatures of 45°C+. Heatstroke, cardiac events, and chronic-disease management are constant — the Ministry of Health needs predictive triage, not reactive medicine.
The pilgrim experience is sacred. Computer vision, biometric flow, and analytics must operate under explicit privacy postures — on-device inference, no facial retention, and transparent governance.
The Kingdom is on a path from ~14M to 30M Umrah pilgrims a year. Doubling throughput without doubling infrastructure means the operating model — not the buildings — must change.
A connected operating picture, not isolated point solutions. Vision into agentic into cloud — wired to the operations center.
Real-time density, flow-direction, and incident detection across 2,400+ cameras. On-device inference; no facial retention by policy.
A 14-language conversational assistant for wayfinding, ritual guidance, lost-companion search, and access to MoHU services on-the-spot.
Forecasting for transit loads, medical demand, and bottleneck risk — feeding into the operations center on a 15-minute horizon.
Sovereign, in-Kingdom landing zone engineered for the 5-day peak. Auto-scaling, edge POPs at the Haramain, and zero-downtime windows.
In-Kingdom SOC, threat hunting tuned to the season, and incident-response retainer with hourly SLA through the Hajj window.
A unified common operating picture for the Haramain — sensors, agencies, and pilgrim flow in a single decision-support surface.
From visa OCR to wearable health monitoring — the most comprehensive Arabic-first AI catalog purpose-built for the pilgrim journey, every stage from intent to post-trip.
24/7 multilingual conversational assistant for pilgrim queries across WhatsApp, web, and app — Arabic-first, ritual-aware.
Middleware connecting operator booking systems to the Nusuk platform — automated registration and status sync.
Personalized packing lists and pre-trip checklists by pilgrim nationality, season, health, and package type.
AI-optimized itinerary planner for Makkah/Madinah heritage sites with real-time transport and crowd data.
Operations dashboard for operators showing pilgrim satisfaction, incidents, and service performance in real time.
Mobile app tracking each pilgrim's progress through the rituals with BLE-based location verification.
AI-powered analysis of reviews, surveys, and social media in 20+ languages to identify service gaps.
Real-time GPS fleet tracking, driver fatigue monitoring, and predictive maintenance for pilgrim transport.
Demand forecasting and menu optimization for hotels and group catering — reducing waste, honoring dietary needs.
AI shopping companion helping pilgrims find halal gifts, compare prices, and navigate Makkah/Madinah souks.
High-throughput Arabic OCR with face matching and Nusuk auto-population for the visa workflow.
Voice + touch kiosks at airports, stations, and Haram vicinity in 20+ languages with offline capability.
Dynamic route optimization for peak-season pilgrim transport using live traffic, sensors, and GPS feeds.
ML models predicting transport and service demand 7 days ahead, aligned to Islamic calendar patterns.
IoT sensor network for air quality, temperature, humidity, noise, and CO₂ across Haram zones with anomaly detection.
Personalized AI tutor teaching rituals by madhab in the pilgrim's native language with 3D Haram visualization.
Photogrammetric 3D model of Masjid al-Haram with guided tours, AR wayfinding, and accessibility routes.
BLE wristband + app system reducing reunification time from hours to minutes with geofencing and meeting-point guidance.
Indoor navigation for visually impaired, wheelchair, and hearing-impaired pilgrims with voice and haptic guidance.
Computer vision detecting collapses, crowd-crush risk, and heat stroke from existing CCTV with auto-dispatch.
Dynamic risk scoring and OCR-powered auto-claims for Umrah travel insurance — settling claims in days, not weeks.
AI layer over Oracle Opera PMS — predictive housekeeping, group room optimization, multilingual concierge.
AR-powered Madinah heritage guide with scholar-authenticated content and narration in 30+ languages.
CCTV-based crowd density mapping for the Prophet's Mosque with Rawdah token system and 30-minute congestion forecasts.
Photogrammetry + LiDAR archival platform creating digital twins of heritage sites with manuscript cross-referencing.
AI health platform ingesting wearable data to detect heat stroke, cardiac events, and fatigue 15-30 minutes before emergencies.
Two engagements that anchor our Haramain practice. Names redacted under MNDA — the Ministry knows the work.
A fused-vision pipeline operating across 380 cameras around the Mataf and Mas'a, producing one-meter density grids and flow-direction vectors at one-second cadence. Wired into the GCMK operations center so coordinators can pre-empt — not react to — pressure points.
An agentic Arabic-first assistant supporting 14 languages, surfaced inside the pilgrim's existing app. Answers ritual questions, locates lost companions, and triages medical or logistical requests against the right MoHU service queue — all without leaving the app.
Authorities, sovereign infrastructure, and the partners we integrate with at the Haramain.
Computer vision runs on-device or at the edge. Faces are not retained. Crowd intelligence operates on density and flow signatures, not identity. Every deployment ships with a written privacy posture aligned to PDPL and the Ministry's standards.
Yes. Our Hajj-window posture is five-nines, no-maintenance, with a 24×7 incident command in Makkah. We've supported a full Hajj cycle without a single P1 incident.
Today: Arabic, English, Urdu, Indonesian, Turkish, Bahasa Malay, French, Hausa, Persian, Bengali, Hindi, Tagalog, Russian, and Mandarin. Twelve more are on the roadmap, prioritized by pilgrim source-country volume.
SDK and API. We do not displace pilgrim apps — we plug into them. The pilgrim copilot is a capability surfaced inside the experience the Ministry already owns.
Multi-year mission programs with a Saudi-onshore team. We run a permanent Makkah operations cell during peak season and a year-round Riyadh engineering base.
Doubling pilgrim throughput is fundamentally an operating-model problem, not an infrastructure problem. Our Hajj practice is built around throughput-per-square-meter improvements — flow design, predictive ops, and pilgrim self-service — that scale before the buildings do.
Sixty-minute working session with our Hajj & Umrah lead and a vision engineer. Bring the bottleneck, the heat-event, or the language gap. We'll come back with a one-page operating-model proposal you can take into the Ministry.