KubeFM

Most teams pick one Kubernetes platform and stick with it. Przemysław Wojtunik explains that his team still works across on-premise infrastructure and AKS in Azure, depending on where customers want their data and how they onboard. Watch the full interview: https://ku.bz/TJRYGMWV2

Alessandro, Research Software Engineer @ IBM, explains how his team addressed critical resource contention issues by moving from manual cluster administration to a GitOps-first approach. Alessandro details their strategic goals: implementing a single source of truth for cluster configuration to provide visibility and traceability of admin actions, particularly for resource quota changes that need to be tracked and potentially reverted. He discusses their approach to preventing GPU resource starvation during high-demand periods, especially when researchers are rushing to complete experiments using LLMs and scientific computations. Watch the full episode: https://ku.bz/5sK7BFZ-8

Kubernetes can schedule workloads, but AI infrastructure goes further than placement alone. Mauricio Salatino explains that training models, moving them, and serving users introduces a separate operational layer. That is where he sees ML platform work begin.

Calin, DevOps Engineer, demonstrates how to automate Helm chart documentation using helm-docs. The solution uses a pre-commit hook to generate markdown documentation from value file comments automatically. This approach removes the burden of manual documentation maintenance, making it easier for teams to adopt and maintain the unified Helm chart approach without spending excessive time on documentation. Watch the full episode: https://kube.fmhttps://ku.bz/mcPtH5395

A special episode on KubeFM, and a slightly different subject than usual. Kelsey, Eric Abercrombie, and Julius Payne II join Bart to explore what hip-hop can teach us about Kubernetes and how music, creativity, and lived experience shape how we think about technology. You will learn: - Why fundamentals, patience, and repetition still matter more than shortcuts - How Kubernetes, community, and confidence intersect for people entering cloud-native work - What hip-hop, production, and storytelling can teach us about ownership, authenticity, and finding your voice Watch (or listen to) it here: https://ku.bz/czrCCXSLt 🌟 This episode is brought to you by LearnKube — get started on your Kubernetes journey through comprehensive online, in-person, or remote training: https://lnkd.in/ge4aYVZq With 🎙Bart

Phil, GMSVP of Tintri, discusses why organizations should focus on maximizing existing hardware rather than constantly adding more resources. He explains how modern hardware delivers billions of IOPS per second but remains vastly underutilized due to inefficient code and methodologies. He emphasizes that customers often buy "ridiculous amounts of compute" they'll never touch due to functional limitations or latency issues. The solution lies in making code more efficient and using intelligent platforms that can effectively leverage even older technology, ultimately delivering significant cost savings while achieving better performance through maximum duty cycle utilization. Watch the full interview: https://ku.bz/pJP25dzCg

Yue Yin, Software Engineer at ByteDance, explains why they decided to open-source their high-performance scheduler. After battle-testing Gödel in production environments handling 5,000 pods per second, and publishing research at SoCC, significant industry interest in the scheduler's design prompted them to share it with the community. The decision was also driven by ByteDance's desire to give back to the open source community that helped them build their infrastructure. Watch the full episode: https://ku.bz/lMpNng_33

The Kubernetes control plane is where the cluster accepts changes, stores the desired state, and decides what happens next. In this series of articles, you will learn: - How the API server handles authentication, authorization, admission, and storage - How etcd stores the cluster state and why it can become a bottleneck at scale - How the controller manager turns intent into actions through reconciliation loops - How the scheduler filters and ranks nodes before placing Pods https://lnkd.in/gzEEd-eJ 🌟 If you want to level up your Kubernetes knowledge, the next LearnKube training starts this Thursday: https://lnkd.in/ge4aYVZq