My Road to KubeCon series is a wrap!

Over the past two months, I've been covering the latest developments in the cloud-native ecosystem in the lead-up to KubeCon North America 2026. Eight editions in total!

It's been fun to have a weekly column, a recurring deadline, and an opportunity to round up stories on a beat for a bit. I don't think I've ever had a recurring weekly column before.

We covered a lot of ground. There were significant changes to Kubernetes itself, from version 1.37 to the transition toward cgroup v2. We also followed how cloud-native infrastructure is adapting to new AI demands, including large-scale inference, GPU utilization, and emerging agent orchestration frameworks.

Observability was another recurring topic, especially the growing interoperability between OpenTelemetry and Prometheus, and new ways to monitor and optimize infrastructure.

If there's one theme that kept coming up, it's that AI is putting new demands on Kubernetes and the surrounding ecosystem. From how resources are allocated to how workloads are secured, monitored, and managed, there's a lot being reconsidered.

In a nutshell: Kubernetes is undergoing a lot of development work to contend with the new demands of agentic AI.

It's been a big pleasure re-syncing with sources, following new case studies and reports, researching emerging trends, and refreshing my Kubernetes knowledge along the way.



A big thank you to editor Nick Lucchesi and The New Stack for inviting me to write this weekly column, to HPE for sponsoring, and to CNCF for helping amplify it.

Things are changing fast in this space, and I will likely always feel like I know so little. 😵‍💫 But that's also what makes this beat so interesting.

I'm glad I had the opportunity to dig deeper into the cloud-native world and hopefully help readers make sense of some of these shifts in the process.

Now, I'm excited to join the community at KubeCon in Salt Lake City to continue these conversations in person!