Core Safety Pack
PPE Compliance, Zone Intrusion, Line Crossing, Dwell Violation, Loitering, and Occupancy / Crowding.
Platform
Warehouse safety packs on the cameras you already run
Edgentik is now organized around focused policy packs, profile-aware camera deployment, and the same operational loop for incident review, diagnostics, rollout, and proof.
Edge-first by design: Core Safety, Worker Protection, and Forklift Safety all use existing cameras, while the current forklift launch scope stays explicitly bounded to Tier A policies.
Policy Packs
The product now starts with one focused pack
This is not a flat feature library. Each deployment starts with one pack, one bounded scope, and a clear review standard before expansion.
Worker Protection Pack
Worker Down and Worker Fall for direct worker-risk events that need faster response and evidence-backed review.
Forklift Safety Pack
Pedestrian in Forklift Zone, Forklift in Pedestrian Zone, Forklift in Restricted Area, Forklift Line Crossing, and Forklift Dwell in Prohibited Area.
Camera Profiles
Use the right camera profile for the job
Not every camera should run the same runtime path. Packs stay practical because camera profiles make the intended model stack explicit per view.
Core Safety Camera
General detection and tracking, with optional PPE support for foundational zone, line, dwell, and occupancy policies.
Worker Protection Camera
General detection and tracking plus pose-aware runtime for Worker Down and Worker Fall.
Forklift Safety Camera
Person and forklift detection and tracking on the shared enforcement path. This is Profile C in the current launch plan.
Premium / Advanced Camera
Reserved for later high-complexity packs rather than folded into launch claims before validation is complete.
What’s live now
What’s live in the platform today
Shared runtime path for person and forklift entities
Generalized incident actor model
Tier A forklift policies wired through the shared enforcement path
Profile assignment in API and UI
Per-camera profile persistence
Contract-enforced Profile C definition and launch scope
Launch boundaries
Current forklift launch scope is zone, line, and dwell based. Advanced cabin-state, load-state, stop or yield, speeding, near miss, and other vehicle-attribute policies are not part of launch.
Final launch signoff still depends on replay and performance validation.
Operating loop
Connect, review, act, complete
From the feeds you already use to validated incidents, clear review, follow-through, then proof and sign-off. Packs and profiles change the product shape, but not the core operating loop.
Connect
Bring feeds, devices, and site scope into one place.
Review
Turn signals into incidents with evidence and an owner.
Act
Assign action and follow through to closure.
Complete
Handover, proof, readiness, and closeout.
Product proof
One operating view—in the surfaces teams use every day
Incident review, diagnostics, rollout, and readiness stay in one place, so buyers can inspect the work the team will actually do instead of a tour of the stack.
Mission Control
Site status and rollout visibility in one place—across sites when you need it.
Incident Queue
Validated incidents with evidence, an owner, and a clear next step.
Diagnostics
Catch drift and recovery needs before small problems spread.
Acceptance / Readiness
Readiness and acceptance views for a clear next step after the first site proves out.
Why this matters
Built to run operations, not just expose detections
Policy-led packaging
Customers buy a bounded pack, not a vague feature matrix with no clear first deployment shape.
Profile-aware deployment
Each camera can run the right runtime path instead of one overloaded stack everywhere.
Validated incidents, not alert floods
Work still starts with incidents your team can review, not a raw feed nobody owns.
Proof and rollout review
Evidence, readiness, and sign-off stay tied to one operating record instead of a demo artifact.
The win is not just sharper alerts. It is a product shape that matches how teams actually deploy, review, and decide what to expand.
Supporting detail
How the story hangs together
Architecture supports the product. It does not replace the pitch. Cameras stay put, packs stay bounded, and review and rollout live where your people work.
1
Existing cameras
Start from the feeds you already run. Profile choice and policy scope stay explicit per camera.
2
On-site processing
Catch defined risks at the site and open incidents locally, including the current forklift Tier A path.
3
Review and site operations
One place for triage, health, rollout, and profile-aware configuration, so the floor is not juggling three tools.
4
Optional handoff downstream
When you’re ready, status and evidence can flow into tickets or other systems, shaped with your team and your launch boundary.
Policies, profiles, and rollouts stay aligned from the floor to the people reviewing.
Integration and handoff · case by case
When your stack needs a continuation
Secondary to the core product: handoff is available when needed, shaped case by case—incidents, acknowledgements, evidence references, and workflow state—not the primary reason teams buy Edgentik.
Webhook or polling handoff
Optional push or pull patterns—scoped to what your process actually needs.
WMS / ERP / ticket continuation
Downstream continuation when you’re ready; breadth and timing decided together—not a one-size integration mandate.
Reviewer acknowledgements
Carry ownership and next steps into the follow-through your site already runs.
Sign-off and proof artifacts
Stakeholder-ready outputs tied to the same record—when proof needs to leave the app.
Go deeper
Rollout, trust, and first policy packs
Start with how a first site goes live, how trust is bounded, or which pack is the right launch point.
See the First-Site Rollout
From readiness through live operation to final review—step by step.
Review Trust Controls
Access, data handling, and recovery—stated plainly for buyers and IT.
See Policy Packs
Core Safety, Worker Protection, and Forklift Safety, with clear launch scope and what comes later.