Outcomes · For engineering teams
Agent workloads kept in the UK
Your agents' compute and run records are planned to live in a UK region, so you can say where they are.
Who it is for
Who this is for
Engineering, security and compliance teams who need to know where agent compute and run records live before a workload is approved.
What changes
Before and after
Before
- Agents run wherever the tooling happens to put them.
- Nobody can state plainly where the compute, logs and run records are held.
After
- Planned: agent environments launched in a UK region.
- Planned: the region, providers, log storage and retention confirmed in writing before any pilot.
How it is delivered
Delivered by a product, not a project
To be delivered by Ithaca, which is in development. UK-hosted compute is the planned position, subject to provider validation.
- The initial scope is Linux CPU environments.
- The provider, isolation runtime, storage, backups, retention and support access must be confirmed before customer workloads are accepted.
What you provide
What we need from you
- The workload you want to run, and every system and credential it needs.
- Your region, data classification, retention and isolation requirements.
- Expected usage and machine requirements.
How the result is measured
How you will know it worked
- Nothing can be measured yet. Before a pilot, the region, providers, log storage and retention are to be confirmed and documented.
- Isolation, logging coverage, recovery behaviour and data-location commitments must be tested and documented before pilots.
How to start
How to start
Ithaca is in development. The waitlist does not provision compute or guarantee a launch date.
Good to know
Limits, stated plainly
- UK data residency is a launch objective, not an available guarantee. It is not currently an Ithaca service commitment.
- You cannot create a machine today.
- No security certification, SLA or benchmark is claimed.
- GPUs, Windows, macOS and desktop streaming are outside the initial plan.
For engineering teams
Other outcomes
- Agents safe to put into productionEvery tool call your agent makes is checked against permissions before it runs, and the actions that matter wait for a person.
- Agents that finish the job after a failureWhen a worker dies mid-run, the agent resumes where it stopped and reuses the steps it already completed.
- Lower model cost without losing accuracyA smaller open model is fine-tuned on your own examples and measured against your current model, so you switch only if it holds up.