Suspected moto crime · Callao · 15:22:31Real-Time Crime Centers
Verified alerts instead of manual feed monitoring, multi-agency coordination on one map, and a court-ready record of every query.
Read More
Where public safety and accountable AI meet. The platform is one system, but the problem changes with the estate, and these are the verticals IREX runs in production today.
The Pattern
A metro platform, a border crossing, and a school corridor need different detectors, different camera placement, and different escalation paths. They do not need different platforms, and buying three is how agencies end up with three systems that cannot talk to each other.
IREX ships the full video analytics module catalog together and lets the deployment select what a site needs. What changes per vertical is the camera specification, the watchlists and integrations, and who receives an alert. What does not change is the accountability: a mandatory Case ID and a non-erasable audit trail, everywhere.
The pages below describe production deployments, not target markets. Where a claim rests on a partner’s reported operational data rather than our own measurement, it says so.
Verticals
Every estate IREX runs in production, one card each. Where a card carries a picture, it is a frame the destination page already shows, returned by a detector on a camera the operator already had, so the picture previews that estate rather than adding a second event. The scenes are illustrative. The transit estates also share an overview of their own.
Suspected moto crime · Callao · 15:22:31Verified alerts instead of manual feed monitoring, multi-agency coordination on one map, and a court-ready record of every query.
Read More →
JDN-529-B, Ford, White · Guadalajara · 11:48:22Watchlists, weapons, vehicles and investigations across jurisdictions, in production in Southern California and Peru.
Read More →
Unattended bag · London · 08:33:04People on the track, train surfing, dangerous crowding at the platform edge and unattended luggage, in the visual conditions underground stations actually have, plus what the same cameras see after the last train.
Read More →
Male, 34, Orange jacket · Dublin · 10:21:48Dublin Airport: watchlist recognition, unattended luggage, weapon detection at access points, and crowd density at peak.
Read More →
Drone launch · Amman · 17:12:38Power stations, oil terminals and pipelines, gas facilities and logistics sites: perimeter, fire and smoke, restricted zones.
Read More →
Underpass flooding · Jakarta · 05:31:44Kazakhstan’s national smart-city program and Peru’s multi-city command centers, on one intelligence layer.
Read More →
Accident / breakdown · Santiago · 09:03:12Mongolia’s nationwide multi-camera enforcement, and 2.3 million violations detected in a single city in one year.
Read More →
Dangerous crowding · São Paulo · 18:52:06Every Premier League national stadium in Bosnia and Herzegovina: banned-fan watchlists and crowd management.
Read More →
Mobile device in use · Plymouth · 14:47:52Continuous monitoring between officer rounds, identity verification at checkpoints, and an audit record that holds up.
Read More →
Boat approaching · Alexandria · 07:29:52Face and vehicle recognition across crossings and ports of entry, with interagency collaboration and full audit logs.
Read More →
Fence line crossing · Nogales · 21:52:16US federal, state and local: CJIS-aligned deployments, vendor identifiers, and the FedRAMP position stated plainly.
Read More →
XPL-263-F, Ford, Silver · San Diego · 03:41:55Organized retail theft and loss prevention, in production across retail chains in Israel and a Florida pharmacy chain.
Read More →A dispersed estate with variable connectivity, where a single edge server per stop is usually the right answer rather than a fiber run.
Read More →Weapon detection, access control and tracking through corridors, on the cameras a school already has.
Read More →Occupancy and lot-full alerts, plate-controlled entry at the barrier, vehicle attributes, and weapon detection across decks and lots.
Read More →Exclusion and self-exclusion watchlists, valet plate recognition, and an audit record on infrastructure the property hosts itself.
Read More →Eight, all production deployments: the United States, the United Kingdom, Ireland, Peru, Israel, Bosnia and Herzegovina, Mongolia, and Kazakhstan. We count nothing else.
Yes, and that is what the multi-tenant architecture is for. Users sit in a hierarchical catalog of groups, each granted specific cameras, watchlists, locations and monitors, with six roles from Viewer to System Manager. It is what lets a national tier and municipal instances share a platform without sharing everything.
You do. Customers retain 100% ownership and control of their data, and IREX neither owns nor accesses customer video, events, logs, watchlists, or floor plans. On-premises deployment is the data-residency mechanism.
With a pilot: one site, three to five use cases, six to twelve weeks, and two or three measurable success criteria agreed in writing before it begins. Accuracy is benchmarked on your own cameras, and the measured numbers go into the contract.
Keep Reading
This page is the index. These are the places it points outward.
Camera count, the sites that matter, and who needs to receive an alert. That is enough for a straight answer about fit.