1
Pull every Datadog alert from the past 24 hours, group by service and severity, and flag anything that fired more than three times without a linked incident ticket.
Persona
IT Ops
AI for IT ops, connected to the tools you actually use.
The stack
A day in the chat
2
Show me all Sentry errors that spiked more than 200% week-over-week and cross-reference them against the GitHub commits deployed in that window.
3
Run an AWS cost audit for this month — show spend by service, flag anything that grew more than 30% versus last month, and draft a one-paragraph summary for finance.
4
Pull every GitHub repo that hasn't had a commit in 90 days and has more than two open security alerts — draft the remediation or archive recommendation for each.
5
List every employee who was offboarded in the past 30 days and check whether their AWS IAM roles and GitHub org access have been fully revoked.
6
Summarize all Slack messages in #incident-response from the past week and tell me how many incidents we had, average time to resolution, and which services were involved.
7
Draft the monthly infra cost report for the engineering leadership meeting — pull AWS spend, highlight the top 3 cost drivers, and include the trend versus the last two months.
8
Check our on-call rotation in Notion and flag if any engineer is scheduled for more than two consecutive weeks of primary on-call duty.
Common actions
Create a Datadog monitor for a new alert threshold
Open a GitHub issue for a security finding
Post an incident update to Slack
Terminate an AWS EC2 instance asks before doing