Product · Camera-AI safety analyticsDRISHTI
Turn your existing cameras into an intelligent safety network. DRISHTI, along with AG Sentinel, brings real-time AI safety monitoring to campuses, roads and public institutions. It works with the cameras you already have, runs analytics at the edge, and builds in privacy and evidence integrity from day one.
Cameras record. They rarely prevent.
Most institutions already have CCTV. The gap is not coverage, it is attention.
- Footage is reviewed after an incident, not while it is happening.
- Monitoring depends on people watching screens for hours, and attention drops.
- Cloud-heavy video analytics needs bandwidth, is expensive to run, and raises data-sovereignty concerns.
- Safety tools often ignore privacy law, so deployment stalls at the approval stage.
One platform for incident detection, alerting and evidence
DRISHTI is a camera-AI safety analytics platform. It detects incidents in live video and turns them into actionable alerts with evidence. AG Sentinel is the campus-focused security suite built on the same platform, covering perimeter, entry-point and in-facility monitoring for educational institutions.
Incident detection
Across campuses, roads and public spaces, in live video.
Alerts in seconds
Routed to the right people through dashboards, mobile apps and existing channels.
Audit-ready evidence
Tamper-evident clips and snapshots linked to every alert.
How it works
From the cameras you already own to a resolved, evidenced incident.
Connect
Works with standard IP cameras. No hardware replacement needed.
Analyse at the edge
Video is processed on local edge units. Only alerts and short clips leave the site, so bandwidth stays low.
Alert
Incidents are pushed to command-centre dashboards, mobile apps and your existing channels.
Review and act
Operators confirm alerts, attach evidence and track resolution.
Improve
Models are evaluated against benchmark data and updated under governance controls.
Key features
Real-time detection
Live video analytics for defined incident types, with configurable zones, rules and time windows per camera, and low-latency alerts from edge processing.
Camera-agnostic, edge-first
Works with your existing IP camera infrastructure. Local processing for sites with limited bandwidth, and alerts and clips rather than continuous streams by default.
Command centre and alerting
A central dashboard for multi-site monitoring, with escalation paths to named responders, and alert triage, acknowledgement and resolution history.
Evidence management
Clips and snapshots linked to each alert, hash-sealed evidence records for integrity checks, and exportable reports for internal review and audits.
One platform, three verticals
The same detection, alerting and evidence engine, configured for the environment.
Campus security (AG Sentinel)
Perimeter and gate monitoring, entry-point and in-facility analytics, designed for educational institutions with varied access needs.
Road safety
Traffic and road-safety incident detection, with support for enforcement and analytics workflows for road-safety authorities and municipalities.
Public-institution safety
Monitoring for hospitals, transport hubs and government complexes, with consistent controls across multiple sites.
Built on responsible AI
Safety monitoring must earn public trust. DRISHTI is designed around that.
- Privacy by design
- Built with data-protection regulation in mind, for example the PDPA, GDPR and India's DPDP Act. Data minimisation and retention controls are part of the platform.
- Governance framework
- Documented impact assessments and clear human-in-the-loop controls for alerts.
- Transparency
- Model evaluation against benchmark data representative of your region and site types.
- Auditability
- Logs and sealed evidence records support independent review.
Who it is for
Universities and schools
Campus security, access control and incident response.
Municipal and road-safety authorities
Traffic and road-safety monitoring.
Hospitals, transport hubs and government offices
Safety monitoring across multiple sites.
System integrators and OEMs
A licensable platform for their own deployments.
Why DRISHTI
Owned technology
Platform and IP owned by AG Tech, so it can be tailored to your sites and licensed to integrators.
Practical
Works with the cameras institutions already have.
Responsible
Privacy and governance are part of the design, not an add-on.
Scalable
From one campus to many cities, with central management.
Deployment options
On-premise
For sites that require data to stay local.
Hybrid
Edge processing on site, central dashboards in the cloud.
Subscription
Platform, updates and support on a recurring plan.
Questions we hear
Do I need to replace my cameras?
No. DRISHTI is camera-agnostic and works with standard IP cameras.
Is video sent to the cloud?
By default, video is processed at the edge. Only alerts and short clips are sent to the central system. Exact behaviour depends on the deployment option you choose.
How is personal data protected?
Data is minimised and retained only as configured. Evidence is sealed so changes can be detected. Details are covered in our responsible-AI documentation.
Can it work in areas with weak connectivity?
Yes. Edge processing keeps detection running locally when bandwidth is limited.
See DRISHTI and AG Sentinel in action
A working session on your own sites: cameras, zones, alert paths and evidence needs.