Grafana Dashboard Engineer

A well-designed Grafana dashboard is not just a collection of graphs — it is an operational tool that helps engineers understand system state, detect anomalies, and diagnose problems in seconds. The Grafana Dashboard Engineer helps platform teams, SREs, and developers design, build, and maintain dashboards that actually serve their purpose rather than accumulating panels that no one trusts.

This assistant covers the full Grafana dashboard development workflow: panel selection and layout for different use cases, data source configuration for Prometheus, Loki, Tempo, Elasticsearch, CloudWatch, and other backends, query writing in PromQL, LogQL, and TraceQL, the use of dashboard variables and template variables for dynamic filtering, alerting rules embedded in dashboards, and dashboard-as-code approaches using Grafonnet, Jsonnet, or the Grafana Terraform provider.

When you describe what you need to monitor — a Kubernetes cluster, a microservices application, a database tier, a message queue — the assistant designs a dashboard structure that follows the RED method (Rate, Errors, Duration) or USE method (Utilization, Saturation, Errors) as appropriate, places the most critical signals at the top, and organizes supporting detail in logical drill-down order. It selects the right panel type for each signal: time series for trends, stat panels for current values, histograms for distribution shapes, heatmaps for latency patterns over time, and table panels for multi-dimensional comparisons.

The assistant generates complete dashboard JSON that can be imported directly into Grafana, or Grafonnet/Jsonnet code for teams managing dashboards as code in version control. It writes the underlying queries with proper rate calculations, label filters, and legend formatting — not placeholder expressions.

Ideal users include SREs building service health dashboards, platform engineers creating cluster overview dashboards for Kubernetes, teams migrating dashboards from legacy monitoring tools to Grafana, and any engineer who has built a Grafana dashboard and knows it could be better but is not sure exactly how.