Operate
Run what you ship — logs, metrics, and traces (Insights), on-call incidents and alert rules (Alerts), and unified cost (FinOps).
Once a service is running, the Operate pillar tells you how it’s doing. Like the rest of the app, these surfaces are scope-aware — pick a service from the top-bar scope picker and every panel narrows to it.
Insights
Logs, metrics, and traces, plus user-built dashboards and an audit log.
Alerts
On-call incidents and the alert rules that raise them.
Cost
Unified FinOps across compute, LLM, storage, seats, and network spend.
How it fits together
- Insights is the telemetry surface: query logs with LogQL, metrics with PromQL, and look up distributed traces — backed by Loki, Mimir, and Tempo. It also hosts user-built dashboards and the organization audit log.
- Alerts is the on-call surface: incidents flow
firing → acknowledged → resolved, raised by alert rules and by external signals (Alertmanager, Datadog, Sentry, PagerDuty). - Cost is the FinOps surface: a unified rollup of every kind of spend — compute, LLM, storage, seats, network — with allocation, budgets, and optimization.
These map directly to the Operate band in the app’s left navigation (/insights, /alerts, /cost).