Engineering role
AI IT Service Manager
An AI IT service manager that reads your tickets and incidents, finds the problems behind them, and writes the SLA report nobody has time for.
Give it your service desk — Jira tickets, incident notes, monitoring alerts, your change log — and it does the ITIL work that slips when the team is busy fighting fires: grouping repeat incidents into problems with a root-cause hypothesis, checking every change request for risk and a rollback plan, measuring performance against your SLAs, and turning resolved tickets into knowledge-base articles so the same question stops coming back.
Things you could ask for
Written the way you would actually say them. The agent plans the steps itself.
- “Read last month’s incidents in Jira and group the ones that look like the same underlying problem. For each group, give the count, the services affected and a root-cause hypothesis we can test.”
- “Review the change requests due for Thursday’s CAB. For each, check it has a risk rating, a test plan and a rollback plan, and list what is missing.”
- “Here are our SLA targets and the ticket export in Google Sheets. Report response and resolution times by priority, the breaches and the near-misses, with the formulas left in the sheet.”
- “Take the twenty most common resolved tickets this quarter and draft a Confluence article for each, written for the person raising the ticket, not the engineer fixing it.”
- “Every Monday, read the weekend’s Sentry and Grafana alerts alongside new tickets and send me what needs an owner before stand-up.”
What you get back
Service-management material you can take into a meeting: problem records with evidence and a root-cause hypothesis, CAB reviews that flag missing risk and rollback detail, SLA reports with breaches and trends, knowledge-base drafts in Confluence or Notion, and a weekly triage summary sent on a schedule.
Tools it works with
A role is a job description; connectors are what let it do the job on your real work instead of on whatever you paste into a chat.
Where it is the wrong tool
- There is no ServiceNow or Freshservice connector. It works with Jira (including Jira Service Management projects) through the Atlassian connector, Zendesk through the Connectors page, and exports from anything else.
- It does not touch your infrastructure. A remote agent has no shell and no access to your servers, so it diagnoses from tickets, alerts and logs you can show it; a person makes the fix.
- It recommends, it does not approve. Change approval, incident priority and SLA commitments stay with your CAB and service owners.
- Root causes are hypotheses until someone confirms them. It shows the evidence for each one so your engineers can rule it in or out quickly.
Starting an agent with this role
- 1Create a Burrak account — any plan works, and your first week is $1.
- 2In the Marketplace, open the Roles tab, find IT Service Manager and choose “Start an agent as IT Service Manager”. That creates a remote agent with the role’s instructions in its system prompt.
- 3Connect the tools it needs — Atlassian (Jira + Confluence), Sentry, Grafana, Google Sheets, Notion, Linear — from Connectors.
- 4Give it a brief. It keeps working in the cloud after you close the tab, and you are only charged while it is actually working.
The IT Service Manager role, in practice
- Do we need to follow ITIL to use it?
- No. It uses ITIL 4 as a checklist of good habits — separating incidents from problems, reviewing changes, measuring against agreed service levels — and fits them to however your team already works. Ask it for the plain-language version if the vocabulary gets in the way.
- How is this different from the Incident Response Commander?
- The Incident Response Commander runs one live incident: the timeline, the updates and the post-mortem. The IT Service Manager works across all of them over time — which incidents are really one problem, which changes are risky, and whether the service is meeting what was promised.
- Can it update tickets itself?
- Through the Atlassian connector it can comment on and update Jira issues when your brief asks it to. Start with comments and drafts, and widen what it may change once you trust its read of your queue.
Capability reference
Summarised from the role’s instructions, which are adapted from the agency-agents collection (AgentLand Contributors), used under the MIT licence.
- Service catalogue design with clear ownership and request fulfilment
- Incident management with classification, escalation and communication
- Problem management with root-cause analysis and a known-error record
- Change management with risk assessment and CAB review
- SLA monitoring, breach reporting and continual service improvement
Your AI assistant is waiting.
Put it to work today.
Put an agent on the work that never needed a human in the first place — the research, the reports, the follow-ups — and take the week back.
$1 for your first week · Cancel anytime · Works on Mac, Windows, iPhone, Android & Web