What Astuse actually does on day one.
These aren’t roadmap promises. Each scenario below is something Astuse runs in production today, with the same core platform, just configured for the segment. Bring your own footage to the demo and we’ll reproduce any of these on your data in 45 minutes.
01
12 hours of CCTV → answers in 15 minutes (insurance fraud review).
A claimant alleging a slip-and-fall at a major retail location. SIU receives 12 hours of multi-camera footage from the day of the incident. Historically: 3 days of analyst review. With Astuse: a single natural-language session.
Inputs
12h CCTV, claimant deposition (PDF), prior claims history
Outputs
Cited timeline, exported clips, redacted bundle for counsel
Tier
Professional or Enterprise
STEP 01
Ingest
Drop 12h of CCTV + the deposition into the case workspace.
STEP 02
Enrich
AI detection, action recognition and cross-camera matching indexed in < 30 min.
STEP 03
Ask
“Show every appearance of the claimant after 14:00, in order.”
STEP 04
Export
Pinned clips, signed bundle, hand-off to counsel.
02
200 hours of body-cam plus 30,000 documents: find the moment in the deposition.
A civil-rights case where defense counsel needs every body-cam moment that contradicts (or confirms) testimony given on page 87 of a deposition. Astuse cross-references modalities natively.
Inputs
200h body-cam, 30k docs, deposition transcripts
Outputs
Cited timeline, deposition cross-reference, redacted exhibits
Tier
Professional · pilots from Investigate (PAYG)
STEP 01
Ingest
Bulk upload via SDK or SharePoint connector.
STEP 02
Index
Three modality-specific indexes built in parallel.
STEP 03
Cross-ask
“Find all video where Officer X is not wearing the badge mentioned on p. 87.”
STEP 04
Build exhibits
Pin clips + cited paragraphs into the case.
03
License-plate cross-camera search for crime analytics.
A municipal force investigating a string of break-ins wants to know every camera in the city network where a partial plate “ABC ___” appeared in the last 14 days, with confidence scoring. Single query, full archive.
Inputs
City-wide RTSP camera feeds, retention archive
Outputs
Geo + time map of matches, ranked by confidence, with clips
Tier
Government & Security · sovereign deployment
STEP 01
Index
Plate detection on every frame at ingest.
STEP 02
Query
“ABC * in the last 14 days, all cameras.”
STEP 03
Rank
Hybrid retrieval + re-ranking by confidence + recency.
STEP 04
Hand-off
Detective gets a map, a timeline and exportable clips.
04
PPE & safety compliance across distributed sites.
A distribution operator with 12 warehouses needs to enforce hard-hat / hi-vis compliance and detect after-hours intrusion without staffing 12 SOC operators. Detection rules + email/Teams alerts + weekly compliance reports.
Inputs
~400 RTSP cameras across 12 sites
Outputs
Live alerts, weekly compliance scorecard, incident clips
Tier
Enterprise + Urban camera SKU (Enterprise band $35)
STEP 01
Connect
RTSP / ONVIF discovery → all 12 sites in 5 days.
STEP 02
Rule Pack
Pre-built PPE + intrusion + loitering detection.
STEP 03
Alert
Per-site shift leads on Teams, GSOC on webhook.
STEP 04
Report
Weekly compliance + incident dashboard exported as PDF.
05
Sovereign deployment with full edge inference.
A federal agency with sites that legally cannot send video off-prem. Astuse runs on a compact edge appliance locally, with only metadata and audit events flowing to the Canadian cloud control plane.
Inputs
On-prem cameras, classified networks, restricted egress
Outputs
Cloud-hosted case management, edge-side inference
Tier
Government & Security · custom Platform contract
STEP 01
Ship
Pre-imaged edge appliances delivered to site.
STEP 02
Air-gap policy
Only metadata/audit egresses; configurable per camera.
STEP 03
Operate
Same dashboard, same investigators, same audit log.
STEP 04
Audit
End-to-end signed audit trail for ATIP / ATI requests.
See the platform in your environment.
Bring your hardest 30 minutes of footage or your messiest case folder. We’ll show you Astuse running on it, live, in a 45-minute working session.