Democratize Observability: Tips and Tricks using Dynatrace Assist

Explore how SREs, software engineers, and architects can use Dynatrace Assist to democratize observability and accelerate root cause analysis—without requiring deep platform expertise. In this hands-on walkthrough, Andreas Grabner demonstrates how Dynatrace Assist uses natural language prompts, agentic AI, and built-in skills to analyze logs, traces, metrics, and Kubernetes workloads—automatically translating intent into optimized DQL (Dynatrace Query Language) queries. Learn how to move from manual analysis to AI-assisted observability workflows with real-time insights and actionable recommendations. What you'll learn How Dynatrace Assist converts natural language into DQL queries Enabling and using agentic AI capabilities in Dynatrace Automating Kubernetes troubleshooting (OOMKills, CPU throttling, pod failures) Identifying log patterns and error clusters automatically Analyzing distributed traces and detecting performance hotspots Running end-to-end root cause analysis across logs, events, and config Generating actionable recommendations and YAML configurations Using Assist skills and integrating with external AI tools (e.g., Copilot) Dynatrace Assist empowers teams to gain deep observability insights faster—making advanced troubleshooting accessible to everyone, not just experts. 🔗 Useful links Explore Dynatrace Assist capabilities in the product documentation: https://docs.dynatrace.com/docs/dynat... Dynatrace for AI GitHub Repository: https://github.com/Dynatrace/dynatrac... Dynatrace dtctl for Agents and Humans:    • Hands‑On dtctl: Use GitHub Copilot & AI Ag...   Explore the scenario on the Dynatrace Playground: https://dt-url.net/devrel-signup-play... Browse all our developer resources: https://dt-url.net/developer-resources 📖 Chapters 📖 00:00 Introduction: Observability without expertise 00:21 Dynatrace Assist overview and capabilities 01:13 Getting started in the Dynatrace Playground 02:03 Enabling agentic AI in Dynatrace Assist 03:39 Context-aware prompts across apps 04:21 Kubernetes troubleshooting with Assist 05:03 Understanding DQL queries generated by AI 06:41 Root cause insights: OOMKills and pod issues 07:27 Forecasting memory and recommendations 08:37 Generating YAML with optimized configurations 11:28 Log analysis: detecting patterns with AI 12:36 Log pattern extraction and root cause hints 14:09 Fixing OpenTelemetry export errors 16:11 Distributed tracing analysis with Assist 17:01 Identifying slow endpoints and performance issues 18:06 Deep dive into trace hotspots 20:20 Democratizing trace analysis with AI 20:25 Root cause analysis across logs and events 22:12 Agentic RCA: multi-source analysis 23:07 Findings: liveness probe and service issues 24:25 SQL and database query performance insights 25:44 Mapping service dependencies to databases 26:42 Key takeaways: Dynatrace Assist capabilities 27:00 Explore skills, dtctl, and future roadmap 28:00 Wrap-up and call to action 🔬 Have a question? Visit our community and connect with our experts and users: https://dynatr.ac/3M0AzM6 👉 Stay connected Facebook →   / dynatrace   Instagram →   / dynatrace   LinkedIn →   / dynatrace   Bluesky → https://bsky.app/profile/dynatrace.com X →   / dynatrace   Twitch →   / dynatrace