- Type
- not stated
- Format
- Structured metric and event streams served over API with standard integrations into control-room tooling, VMS and PSIM: high-frequency per-zone time-series (occupancy, density, directional flow, queue length, wait minutes, dwell) plus discrete timestamped incident records with type, location and confidence, retained in a searchable event archive. Latency is specified as under 2 seconds from camera event to control-room alert, so the primary form is a real-time stream with the archive as the accumulated secondary asset
- Volume
- Potentially high per site and low in aggregate. Per instrumented hub, order 10^7-10^8 metric rows and 10^5-10^6 archived event records annually, because a live CCTV estate typically runs tens to hundreds of cameras and the product derives metrics from all of them rather than a sparse new sensor grid — the reuse-existing-cameras design is precisely what makes per-site volume high. Aggregate volume across the vendor is however bounded by customer count, which a 2-person company plausibly holds in single digits, so total licensable volume is modest today
- Users
- N/A — B2B control-room deployment with no consumer accounts; scale is counted per instrumented hub and per camera or zone, not per user. There is no registered population, and by design the product does not identify individuals
- History
- 2 years
- Update frequency
- not stated
- Growth
- Active but effectively pre-scale — a 2-person founding team with no funding on record. Activity evidence is real and technical rather than promotional: quantified production accuracy of 98.4%, a sub-2-second latency specification, per-camera threshold and restricted-zone configuration, integration support across multiple named control-room and VMS/PSIM ecosystems, and a live demo, all of which indicate working product deployed somewhere rather than a pitch deck. Scale evidence is absent: no headcount beyond two founders, no disclosed funding, no named customer, and multi-sector marketing across transport and events that reads as a young company searching for a wedge market
- Launched
- Est. 2023-2026 — recent vintage. Basis: a 2-person company with no funding record and an organisational profile that has only just appeared in corporate databases is a strong recency signal, and the product's positioning on top of recent commodity computer-vision capability fits the same window. The company's own reference to validated production datasets shows the model has been trained and run somewhere, so it is not day-one, but there is no sign of multi-year market presence. No founding date appears on the retrieved page; treat this as a bounded estimate and confirm with the vendor
- Delivery
- not stated
- Entity mapping
- not stated
- Sample
- not stated
- Point-in-time
- not stated
- Licence
- not stated