01
Assess the work
Start with the use case, recommended model, and expected team demand.
Dedicated hardware / early access
Turn owned hardware into your team’s private AI service.
Private AI on hardware your organisation owns. CaseDesk turns a Mac, workstation, or server into a governed private AI connection. Keep the hardware and local runtime. Give your people one trusted way to use it.
No public IP, inbound firewall rule, or Kubernetes required.
Customer-controlled hardware / CaseDesk-governed connection
Owned by you01
Mac or GPU server
Your local runtime stays on hardware your organisation controls.
Device AgentAuthenticated outbound connection. Plan-bound changes only.
One compatible team connection
Approved model. Health visibility. Controlled maintenance.
01 / The difference
Running a model is not the same as operating a company service.
Ollama runs a model. CaseDesk governs the company service around it. Local runtimes are excellent tools for an individual developer; when a team depends on one machine, the operating questions change: which model is approved, who owns maintenance, whether the device is ready, and how internal tools connect without opening the runtime to the internet.
Direct local runtime
CaseDesk on dedicated hardware
Model process on one machineManual setup and restart responsibility.
Private team connectionCompatible endpoint and a clear operational boundary around the device.
Mutable local configurationChanges depend on the person operating the host.
Reviewed model and runtime planThe Device Agent acts only on an exact owner-approved configuration.
"Is the machine on?"Availability becomes an ad-hoc support question.
Truthful operating stateSee device, runtime, model, and connection readiness before work is assigned.
02 / The path
From one owned machine to one dependable connection.
CaseDesk does not take over the hardware. It provides the approval and connection layer that lets the runtime serve a wider team without public exposure.
02
Claim the device
An owner authorises a Device Agent on the host. The first report is read-only.
03
Approve the exact plan
Review capacity, model identity, runtime changes, and operating limits before anything changes locally.
04
Operate with evidence
Keep the runtime private while the connection reports health and maintenance state.
03 / Best fit
For teams that value ownership and still need accountability.
Your hardware stays yours. The service becomes easier to trust.
CaseDesk does not buy, operate, or pay for your hardware. It gives the local runtime a governed connection, a visible owner approval path, and a route to repeatable operations as demand grows.
Private coding assistance
Give internal developer tools an approved connection to a local model without exposing the host.
Workloads with a clear operating owner
Use hardware you already own when utilisation, device support, and availability expectations are understood.
Own the hardware.
Operate the connection with intent.
Talk through your workload, device ownership, operating boundary, and the path to a governed private AI service.
Talk to the CaseDesk team