Operations | Monitoring | ITSM | DevOps | Cloud

In short, in the five weeks to 27th October 2026, Okta, Satya Nadella, the Cloud Security Alliance and Salesforce each put the words “kill switch” or “emergency brake” next to AI agents, and the AI Kill Switch Act, introduced in the US House on 23rd July, would require the largest model providers to be able to throttle, suspend or shut their systems down on order. Every one of those switches stops something, and none of them stops the same thing.

October was a big month for Civo, with Starling and Behavox both choosing UK sovereign infrastructure for their AI and Civo Indonesia Region going live in Jakarta. Here is everything you might have missed.

Railway, Northflank and seven other developer platforms disclosed breaches in October 2026. Here is why a shared PaaS turns one breach into everyone's breach, how Qovery gives you git-push deploys while your apps and data run in your own AWS, GCP, Azure or Scaleway account, and what Qovery still holds. Romaric founded Qovery to make Kubernetes accessible to every engineering team. He writes about platform strategy, developer experience, and the future of cloud infrastructure.

Replay and key takes from our October 20 live with TPC: why Day Two is 80% of infrastructure work, why shared staging breaks when agents write the code, how agents cut migrations from months to hours, and how to keep agents under control with policy tokens and audit logs. Romaric founded Qovery to make Kubernetes accessible to every engineering team. He writes about platform strategy, developer experience, and the future of cloud infrastructure.

A tricky Google Cloud support ticket that spanned GKE, Private Service Connect (PSC), Networking, Hostnet, Kernel, and the Linux kernel itself.

An agent that works on your laptop is borrowing three things it doesn’t own: your identity, your attention, and your judgment. It runs with your credentials and your access. It runs when you trigger it. And when it gets something wrong, you notice and tell it, which is the only evaluation it ever gets. Deploy the same skill to the cloud, unattended and pointed at production, and all three disappear at once.

Two-thirds of UK IT leaders say they would switch cloud providers to take back control. Fifteen percent have actually done it. That is the headline from Civo's Digital Sovereignty Revolution 2026, a survey of 1,000 UK IT leaders. The intent is loud and getting louder. The movement is almost silent. It would be easy to call that apathy. It is not. UK organizations are staying on hyperscale infrastructure they no longer fully trust for four very specific reasons. Three of them are structural.

BYOK on EKS, GKE and Kapsule now runs through the Qovery Operator: install it once with Helm on a cluster you already run, and deploy Qovery's component updates from the Console. Now in Preview, for clusters without cert-manager, ExternalDNS or Envoy Gateway. Pierre is a staff engineer on the Qovery platform team. He works on the Qovery Operator and the platform catalog that install and update Qovery on Kubernetes clusters.

Running an LLM on Kubernetes is rarely just a Deployment with vLLM in it anymore. A whole layer of frameworks and operators has grown up for it: KServe, NVIDIA Dynamo, Ray Serve, Kthena, llm-d. Each one adds its own custom resource that describes a model deployment, with routing, autoscaling and the split of a model across pods built in, and each one turns that resource into ordinary Kubernetes objects in its own way.