AI
Read every DevOps talk
in minutes.
TL;DR, full written analysis, and word-level YouTube timestamps for every talk — across every conference we cover.
Browse by topic · all topics →
Conferences
When OTTL Goes Off the Rails: Debugging Transformations with Confidence - Edmo Vamerlatti Costa, Elastic & Tyler Helmuth, Honeycomb
Demonstrated how to debug the OpenTelemetry Transformation Language (OTTL) using three tools: the debug exporter, debug logs, and the OTTL playground. Four live scenarios walked through common failure modes. Case-sensitive pattern matching, JSON cache key misuse, shared resour...
Read the analysis →Recently added · latest analyses across every conference
Visualizing GitOps: A Tour of Flux UIs in the Open Source Ecosystem - Stefan Prodan, ControlPlane
Flux 2.8 ships Helm v4 support with Kubernetes server-side apply, cutting cluster queries and enabling hundreds of parallel Helm releases. A new upgrade-canc...
Unleashing Event Driven Capabilities With KEDA - Jorge Turrado, SCRM Lidl International Hub & Zbynek Roubalik, Kedify
Jorge Turrado and Zbynek Roubalik, both KEDA maintainers, demonstrated live RabbitMQ-driven autoscaling and walked through KEDA's two core resources: ScaledO...
Scaling the Kubernetes Ecosystem: Uber's Approach to Cluster Lifecycle Management - Silvio Simunic & Vadim Plakhtinskii, Uber
Uber's Compute team replaced a 15-step manual cluster provisioning process with a Kubernetes-native operator that turns up, reconfigures, and decommissions c...
A Tale of Two KEPs: How the Community is Taming Kubernetes' CrashLoopBackoff - Yang Li, Google
Yang Li demonstrates how two Kubernetes KEPs, KEP-4603 and KEP-5593, finally address CrashLoopBackoff's hardcoded 300-second maximum delay, introduced in 201...
Spin-Up, Test, Tear-Down: How Trivago Runs Developer Preview Environments at Scale - Armin Aminian, Trivago & Jan Wozniak, Kedify
Trivago runs 200+ active PRs daily across frontend and backend services, each automatically triggering an isolated preview environment on Kubernetes. The set...
Making Topology-Aware Scheduling Practical for AI Workloads: From Discovery to Simulation at Scale - Weizhou Lan, Daocloud
Weizhou Lan demonstrates why static node affinity fails for AI scheduling at scale, and shows how Volcano, Kueue, and Kai Scheduler enforce hierarchical topo...