Sponsored Keynote: The Work Before the Magic: Autoremediation Readiness
summary
Autoremediation succeeds only with all the data (metrics, logs, traces ingested via OpenTelemetry), structured context (normalized log fields, service maps, trace/span IDs, and external signals like recent code commits or data‑center weather events), and tight guardrails: start with low‑risk repeating fixes (cache clears, resource bumps), require easy rollback, and enforce human-in-the-loop approval buttons that convert to automatic actions after repeated confirmations, progressively expanding scope as trust proves the fixes.
AI-written summary · about
related talks
05mKeynote - Before You Let AI Touch Production: Autore...
51mRemediation-Aware Reachability: Patching Containers, Prioritizing with Agentic-CTI, and Scaling...
56mWhy Auto-Fixing Vulnerable Code Needs More Than Good Intentions
27mKickoff + Keynote
41mLessons from Mission-Critical Systems in Fast-Moving Industries
14mFrom Dashboard to Defense: Automating Resilience at
39mVirtuous Cycles: Enabling SRE via automated feedback loops
25mLLM Agents for Site Reliability