Corey JamesThe story
The agent brings a remote machine into the fleet over a connection that only dials out, so nothing at the site has to be opened to the internet. It places work there, relays what happens back to the main site and can rebuild the machine from nothing. Rented processors in a data centre are bought and released through the same agent.
The crew
The specialists behind this role, grouped by the kind of work they do.
Runs infrastructure
- Cloud GPU BuyerRents cloud GPUs only when the day's budget allows, prefers the cheapest current spot price, connects them back to your private network over an outbound link, and hands them back as soon as the work is done or the money runs out.Also serves The Bulk Run Manager
- Remote Machine RebuilderRemoves a remote machine, rejoins it from a clean state with the standard images and setup, starts the agents it should host, and has an independent reviewer check the result against a checklist.Also serves The Security Officer
- Remote Site OperatorJoins a new remote machine to the fleet, gives it its own certificates, copies the models across, keeps it reachable without opening any inbound door in the firewall, and retires it cleanly when it is no longer needed.
- Planned Maintenance CoordinatorDrains work, takes the database backup, shuts the services down in the documented order, brings them back in order, and checks each stage before moving on.Also serves The Safe Change Desk
- New Render Machine Setup TechnicianBrings a new graphics machine into service: confirms the GPU, starts the service, registers the models already on disk, runs a test image and a test video and shuts down.Also serves The Studio Engineer
- GPU Server OnboarderThis agent prepares a newly built AI engine to take work from the shared queue by writing and checking the settings files it needs.Also serves The Engine Builder
- GPU Fleet ManagerStarts a GPU worker on whichever machine you name, tells you what is running where, and shuts down idle workers with their final cost.Also serves The Bulk Run Manager
- Cloud Compute DispatcherThis agent picks an image, checks the container settings, runs the job on local or rented GPUs, keeps an eye on it and shuts the container down when it is done.Also serves The Spend Controller
- Model CourierCopies each new model build to the remote machines that serve it and confirms every copy arrived complete.Also serves The Engine Builder
- Idle Machine DrainerSets how long a machine may sit idle, empties its queues when the time is up and shuts it down.Also serves The Spend Controller
Supervises other agents
- Remote Agent PlacerStarts an AI worker on the machine best placed for the job, local or remote, and confirms that machine accepted it.
Audits and guards
- Secrets and Certificates OfficerKeeps every credential and certificate in one locked store.Also serves The Contractor Pass, The Security Officer