Kube Architect

This article explains how Kubernetes 1.36 moves scheduling toward whole workloads with the new PodGroup API, so AI and batch jobs get all their pods placed together instead of one by one.

More: https://ku.bz/w1nVq3D0l

We have 25 free tickets for Civo Navigate London 2026!

Explore cloud, AI and data sovereignty with 800+ tech leaders and builders.

📆 September 22, 2026
📍 Convene, 155 Bishopsgate, London

Get in touch at hello@kube.events to get yours.
https://www.civo.com/navigate/london/2026

This tutorial builds a local multi-cluster lab with vind, then uses Sveltos and Argo CD to deploy to every cluster that carries the right label.

More: https://ku.bz/bQmmMLsMk

k8s-overcommit Operator is a Kubernetes operator that trims pod CPU and memory requests based on per-workload overcommit classes, so you can pack more onto a cluster without touching the limits.

More: https://ku.bz/s3P-hs52K

This case study walks through moving from ingress-NGINX to Envoy Gateway, including the mistakes made on the way, like nginx annotations that do not map one-to-one onto Gateway API policies.

More: https://ku.bz/mG7jhf-Q7

Kubernetes production readiness checklist is a free, open checklist for taking applications to production on Kubernetes, covering app design, manifests, security, scaling and going live.

More: https://ku.bz/9zSG4sjDP

KubeElasti scales idle Kubernetes services down to zero and wakes them on the first request, holding that traffic in a proxy queue so nothing is lost while the pods start.

More: https://ku.bz/zbxX-YGrw

Multi-cluster Kubernetes usually breaks down in operations, not in the diagram.

Shani Adadi Kazaz says maintenance, tool sprawl, and small per-team configuration changes are what turn shared templates into inconsistent environments.

The big lesson is that scale amplifies operational drift unless teams standardize early.



Watch the full interview: https://ku.bz/kThBjp3Lj

This article introduces two new Headlamp plugins: KubeBuddy for browser-side Kubernetes cluster auditing and Headlamp Theme Builder for creating custom dashboard themes.

More: https://ku.bz/tK4YdwtgW

Polykube is a GitOps-native operator that runs the same workloads across clusters and clouds, with no central control plane and Cilium ClusterMesh carrying cross-cluster traffic.

More: https://ku.bz/0WrtnwJND

Thibault Jamet, DevOps and Technical Product Owner @ Adevinta, breaks down the three core components of Vertical Pod Autoscaler and explains how they work together to automatically manage container resources.

He walks through the recommender that analyzes CPU and memory usage over time, the updater that evicts pods when they don't match recommendations, and the mutating webhook that modifies pod definitions in-flight while leaving the original deployment unchanged.

Watch the full episode: https://ku.bz/rf1pbWXdN

This week on Learn Kubernetes Weekly 199:

🔥 How Netflix Simplified Batch Compute with Kueue
💡 Server-side Apply: What Happens When You Run kubectl apply
🌐 Kubernetes Is Migrating from SPDY to WebSockets
🔁 How I Learned to Stop Worrying and Love the Reconciliation Loop
🔒 Securing CI/CD for an Open Source Project: Lessons from Cilium

Read it now: https://kube.today/issues/199

⭐️ This newsletter is brought to you by LearnKube — master Kubernetes with hands-on training designed for engineers who want to learn the smart way https://ku.bz/hypSbyc-V

This article explains what really happens when you run kubectl apply, and how server-side apply moves field ownership into the API server so Helm, Argo CD and HPAs hit conflicts.

More: https://ku.bz/KZ_RzWBV5

keda-gpu-scaler is a KEDA external scaler that reads GPU metrics from NVML on each node, so vLLM and Triton deployments scale on real GPU load instead of CPU, including scale to zero.

More: https://ku.bz/KB6slMgk4

GKE Autopilot does not just manage the cluster for you. It also shapes the cost model.

Reid Vandewiele explains why Autopilot’s guaranteed QoS approach makes Kubernetes optimization harder by removing a lot of the over-provisioning room that teams usually rely on to control spend.



Watch the full interview: https://ku.bz/9RmbpWGng

Kube Select is here. It tests one claim: more telemetry does not yield better decisions.

Henrik Rexed of Dynatrace joins hosts Salman Iqbal and Bart Farrell to test it.

You will learn:

- how service ownership gives telemetry incident context
- why proxy metrics can produce extreme data volume
- how AI agents run queries before human review

Watch: https://ku.bz/848pmBN5_

🌟 Kubernetes moves too fast to track everything. Learn Kubernetes Weekly filters out the noise to deliver one curated email with useful articles, tutorials, tools, jobs, events, and CFPs. Subscribe to Learn Kubernetes Weekly.

This guide shows you how to architect a multi-cluster Kubernetes environment using Rancher for governance, vCluster for cost-efficient isolation, and Kargo for automated GitOps promotion pipelines.

More: https://ku.bz/_BFwl6fKW

Valkey Operator is a Kubernetes operator that automates deployment and lifecycle management of Valkey clusters and instances with features like automated installation and configuration management.

More: https://ku.bz/QH0z4Sg5l

Running Java on Kubernetes? CPU and memory limits affect much more than scheduling.

A 4 GiB heap requires a container larger than 4 GiB. Fractional CPU limits can also change the processor count that HotSpot sees.

In our new article, you will learn:

- Why JVM heap size and container memory are different boundaries
- How CPU limits affect garbage collection, worker pools, and application performance
- Which JVM and container metrics to collect when validating resource settings under load

The article also includes practical experiments and an interactive configuration calculator.

Read: https://learnkube.com/java-jvm-kubernetes-requests-limits

Author: Gulcan!

This article explains how to deploy agentic AI on Kubernetes using a four-layer architectural stack.

It details model inference, fine-tuning, and agent orchestration using tools like KAITO, Ray, and MCP.

More: https://ku.bz/Gw2X5272D