Managed operations

Get engineers to help keep your software running.

Our engineers use Operate to investigate problems and look after the systems you agree with us. Your team stays involved in priorities and approvals, while we handle the agreed day-to-day work.

Discuss your scope Read the documentation

Agree what we will look after

Choose the applications and services you need help with. We agree the work, access permissions, working hours and who to contact when something needs attention.

Help us understand your systems

We set up Operate, connect the information it needs and go through the first findings with your team. If an internal system needs its own connection or checks, we can build them as part of the agreed work.

Handle problems and the follow-up work

Our engineers check what Operate finds, investigate problems and help your team make the agreed changes. The explanations stay in the knowledge library so they can be used again.

Agree who does what.

Before starting, we agree how much engineering help you need, when our team is available and how decisions and changes are approved.

Named delivery contacts and escalation paths

Agreed systems and engineering capacity

Coverage and response expectations

Review and approval responsibilities

Model usage and deployment costs

Know who to call
and who does what.

Before we start, we agree which systems we will look after, what access we need and who approves changes. The table below shows how the work can be divided.

Production responsibilityOur engineers + OperateYour team

Monitoring and review

Check the findings for the systems we agreed to review.

Choose priorities and approve access to your systems.

Investigations

Investigate what went wrong and suggest a fix.

Explain business rules and tell us who can help.

Production changes

Prepare code changes and help apply them where agreed.

Approve changes through your normal review process.

Knowledge and follow-through

Keep the explanation and complete the agreed follow-up work.

Decide priorities with our engineers.

What a week might look like.

Our engineers might investigate a failed checkout, check why an orders page is getting slower and prepare a database change for your team to approve. They would also keep what was learned in the knowledge library. The actual work depends on what you agree with us.

Trust and control

You decide what Operate can access.

Operate needs information from your systems to investigate a problem. You choose what it can read, where it runs and which AI model it uses.

Run it on your own servers
Install Operate with Docker. Use your own AI provider or a model running inside your network.
Choose what it can read
Give it accounts with access only to the information it needs. Its queries are recorded.
Review changes before applying them
Operate suggests a code or configuration change. Your engineers check it and apply it.
Check where the answer came from
Open the records, logs and code used in the investigation.
See what was checked
The report includes explanations that were considered and ruled out.
Look back at the investigation
The steps are recorded so your team can review how the answer was reached.