I Stopped Staring at Dashboards. AI Reads My Grafana Metrics Now.

Why It Matters

AI‑augmented observability accelerates incident diagnosis and remediation, cutting operational costs and improving system reliability.

Key Takeaways

• Grafana Assistant lets AI query metrics, logs, traces via chat.
• AI can generate custom dashboards instantly from natural language prompts.
• Cloud Code integrates with Grafana MCP for terminal‑based observability analysis.
• AI analysis works in both hosted Grafana Cloud and self‑managed setups.
• Automated dashboards reduce incident response time compared to static panels.

Summary

The video introduces Grafana Assistant, an AI‑powered agent embedded in Grafana that can read metrics, logs, and traces directly from a chat interface. It also shows how Cloud Code can talk to the Grafana MCP server, letting developers stay in their terminal while querying observability data.

Key insights include the assistant’s ability to infer the correct data source, build PromQL/LogQL queries, and render results as panels or textual summaries. It can also generate full dashboard JSON on demand, copy existing dashboards, and even attempt to modify legacy JSON formats, all without opening a browser. The demo covers CPU/memory queries, log extraction that surfaces hidden issues, and trace listings that highlight instrumentation gaps.

Notable examples feature a one‑click chat that returns per‑container CPU charts, log panels that automatically flag a stuck reconciliation and a WordPress scan, and a trace view that points out missing instrumentation. The assistant even creates a brand‑new dashboard in seconds, then clones and tries to edit a community kube‑state‑metrics dashboard, exposing the challenges of Grafana’s older JSON schema.

The implications are clear: AI‑driven observability can shave minutes—or even hours—off incident triage by eliminating manual dashboard hunting and query crafting. By unifying analysis and remediation in the terminal, teams reduce context switching, lower operational overhead, and can respond to production alerts faster, positioning AI as a core productivity layer for modern SRE workflows.

Original Description

Modern observability workflows are getting a major upgrade. This video explores how AI agents can now read your metrics, logs, and traces directly — drawing conclusions, building custom dashboards on the spot, and surfacing problems you didn't even know to ask about. Starting with Grafana Assistant inside the Grafana UI, then moving to Claude Code wired up to the Grafana MCP server, you'll see how a single terminal-based agent can query Prometheus, Loki, and Tempo, generate fitted dashboards from natural language, and analyze runtime data without ever switching tools or leaving the command line.

The real payoff isn't prettier dashboards — it's closing the gap between analysis and remediation. When your coding agent can see what's actually happening in production, it stops reasoning in the dark. It verifies fixes by checking live data after a deploy, notices anomalies on the way to doing other work, and grounds every decision in real observability signals. Whether you're using Grafana Cloud's built-in Assistant or connecting an external agent through the MCP server, the result is the same: a code-and-runtime-aware agent that handles the tedious parts of incident investigation so you can focus on what actually matters.

#Grafana #AIObservability #ClaudeCode

Consider joining the channel: https://www.youtube.com/c/devopstoolkit/join

▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬

➡ Transcript and commands: https://devopstoolkit.live/observability/i-stopped-staring-at-dashboards-ai-reads-my-grafana-metrics-now

🔗 Grafana Agent: https://grafana.com/products/cloud/ai-observability

▬▬▬▬▬▬ 💰 Sponsorships 💰 ▬▬▬▬▬▬

If you are interested in sponsoring this channel, please visit https://devopstoolkit.live/sponsor for more information. Alternatively, feel free to contact me over Twitter or LinkedIn (see below).

▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬

➡ BlueSky: https://vfarcic.bsky.social

➡ LinkedIn: https://www.linkedin.com/in/viktorfarcic/

▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬

🎤 Podcast: https://www.devopsparadox.com/

💬 Live streams: https://www.youtube.com/c/DevOpsParadox

▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬

00:00 Observability for AI Agents

01:43 What Is Grafana Cloud and Grafana Assistant?

05:15 Grafana Assistant Demo: Metrics, Logs, and Traces

09:35 Claude Code with the Grafana MCP Server

12:44 Generate Grafana Dashboards From Natural Language

19:24 Grafana AI Observability: My Verdict