Kiosks that reset themselves.
Public kiosks face constant tampering and drift; each session runs statelessly and leaves nothing behind, no residual data, no manual reimaging.
An unattended device, exposed to everyone, all day.
Self-service kiosks sit in public, unsupervised, taking input from anyone who walks up. They get tampered with, broken out of their lockdown, and quietly accumulate session data from every person who used them. When one drifts or is compromised, someone has to drive out and reimage it, and until they do, it's both a security hole and a bad customer experience.
The pressure points in kiosk fleets.
Tampering & breakout
Public users try to escape the kiosk shell into the underlying OS.
Residual session data
Names, payments, and inputs from the last user can linger on the device.
Field reimaging
A drifted or broken kiosk needs an on-site truck roll to fix.
Inconsistent uptime
Distributed kiosks fall out of sync and out of service.
Stateless kiosks, managed from one console.
Leaves nothing behind
Each session runs statelessly and resets, no residual data between users.
No manual reimaging
Tampering or drift is undone by the next clean boot, not a site visit.
Locked to its role
The device runs only the authorized kiosk workload, nothing else.
Update and repurpose remotely
Push changes or change a kiosk's role centrally from Switchboard.
On-site visits to recover a drifted kiosk.
A kiosk that resets itself stays in service without a technician in the field.
Kiosk operations, answered.
What if someone breaks out of the kiosk app?+
There's nowhere persistent to land. The session is stateless and resets to a verified baseline, so any tampering is wiped on the next boot.
Is customer data safe between users?+
Yes. Nothing entered in one session is retained for the next, the device keeps no local state.
Can we repurpose a kiosk later?+
Yes. Reassign its role and policy from Switchboard without reimaging or visiting the site.
“Tamper with it all you want. The next boot is clean, and you were never anywhere persistent.”
See the stateless endpoint on your own hardware.
Flash an idle machine into a live endpoint and run your real workloads. You buy no hardware and sign nothing.