https://shengxu.pages.dev/en/tags/holmesgpt/Cloud architecture & DevOps notes by Shengxu: Kubernetes, Cilium, observability, LLM infra, AI agents.Hugo 0.153.2 & FixIt v0.4.0-alpha.3-20251225101113-8ffb9a95enFri, 17 Apr 2026 19:40:00 +0800https://shengxu.pages.dev/en/posts/azure-sre-agent-to-holmesgpt/Fri, 17 Apr 2026 19:40:00 +0800https://shengxu.pages.dev/en/posts/azure-sre-agent-to-holmesgpt/AIKubernetesDevOpsObservability<p>In the multi-cloud Kubernetes era, the pain point for SREs is no longer just &ldquo;too many alerts,&rdquo; but rather investigation chains that are too long, context that is too scattered, and troubleshooting costs across clouds that are too high. What truly drains people isn&rsquo;t glancing at a chart, but constantly switching between multiple cloud platforms, logging systems, deployment records, and ticketing systems.</p> <p>This is why AI SRE Agents are starting to deliver real value. Their goal isn&rsquo;t to be a better conversational Copilot, but to proactively take over the highly repetitive first half of the work—&ldquo;checking logs, finding correlations, guessing root causes, and giving suggestions&rdquo;—once an alert is triggered.</p>