Operating agreement
A written agreement on what we cover, how issues are escalated, who owns what, what is excluded, and which maintenance we do.
Managed AI and data operations
We monitor and maintain named applications, integrations, and data pipelines. The agreement sets out support hours, response commitments, and what is excluded.
ScopeSpecific improvements or ongoing support for named systems
Delivery timingOngoing, on an agreed schedule
How we check itService reports track the agreed health checks, incidents, and maintenance, and each agreement sets what is covered.
Know who responds when something breaks
Before we take on support, we review the system, its dependencies, and its recent failures. Then we agree which alerts we respond to, what maintenance is included, and when another team needs to be involved.
Good fit
- Teams that want a named partner responsible for keeping a system running
- Organizations without enough staff to maintain the system themselves
- Leaders who want regular reports on system health
Not the right fit
- Teams expecting 24-hour support that isn’t written into the agreement
- Systems we can’t get enough access or documentation to support
- Organizations with no one of their own responsible for product, data, or security decisions
What you get
Monitored services
Checks on system health, data freshness, AI answer quality, and cost for the systems in scope.
Maintenance record
Written procedures, incident notes, a log of scheduled changes, and regular service reviews.
How it works
Review before taking over
Before we take on support, we review the system’s dependencies, access, failure history, and recovery procedures.
Agree on coverage
We agree with you on what we monitor, who responds, and which issues go to other teams.
Operate and review
We carry out the agreed maintenance and use each incident to find changes that make the system more reliable.
Scope, cost, and ownership
What we need from you
- Production access and named business and technical owners
- Escalation contacts and agreed recovery expectations
What affects cost
- Service criticality and required support hours
- Number of pipelines, applications, and providers
- Number of incidents and amount of routine maintenance
Technical scope
- Health checks
- Monitoring
- Scheduled maintenance
- AI quality reviews
- Incident records
- Operational reporting
Support and maintenance
The service agreement lists the monitored systems, support hours, response commitments, and work that needs a separate request.
Common questions
Does this include a public SLA?
No. Service levels, coverage, and response times are agreed for each engagement once we understand the system and what it needs.
Can this follow an initial build?
Yes. It works best once responsibilities, monitoring, and recovery steps are in place.