Kubernetes changes, explained by the people who made them.
Kaslin Fields and Abdel Sghiouar interview the people behind Kubernetes: release leads, SIG maintainers, technical writers and Google Cloud engineers. Topics range from what each new version graduates and deprecates to Gateway API, the documentation and a new AI conformance program.
PRESS PLAY
Kubernetes Podcast from Google covers the Kubernetes project from the inside, through interviews with the people who maintain it, document it and ship it. Kaslin Fields and Abdel Sghiouar host. Some interviews are recorded solo, including two that Kaslin Fields recorded at KubeCon North America 2025 in Atlanta. Episodes often follow a set pattern. News comes first: Knative reaching the CNCF graduated maturity level, GKE Autopilot becoming available inside standard clusters, a GKE anniversary hackathon built around agentic AI, the KubeCon schedule going live. Then the guest takes over. Release episodes are a fixture, since the hosts treat inviting each upcoming version's release lead as a show tradition. In 'Kubernetes 1.35: Timbernetes, with Drew Hagen', the conversation covers the count of enhancements heading to stable, beta and alpha, plus the deprecations. Drew Hagen, a senior DevOps engineer, describes the lead role as delegation, coordination and making the final calls, with release shadows in other time zones often catching issues first. 'Kubernetes 1.34 Of Wind & Will, With Vyom Yadav' features a software engineer from the security team at Canonical. That conversation looks at why an option called 'prefer close' reads as vague next to 'prefer same zone' and 'prefer same node'. It also reassures anyone worried about YAML: a new alpha option is itself YAML, and YAML is staying. Beyond releases, the show spends time with individual SIGs. Antonio Ojea, a core maintainer and a tech lead of SIG Networking and Testing, explains that Ingress is considered frozen, that new features go into Gateway API, and that the project still wants to know what keeps most users on Ingress. The same conversation mentions the Inference Gateway, an extension for better routing to models. Shannon Kularathna, a technical writer, talks with Kaslin Fields about SIG Docs work such as regenerating reference pages for every release, and asks for more help reviewing blog posts. That episode opens with a Reddit joke about crying through the official documentation. Later it drifts into how to say kubectl, and one speaker compares a cuter pronunciation to a crab scuttling across the screen. Google's own managed service gets airtime too. Gari Singh looks back on ten years of GKE and weighs extensibility against a stable core, citing DRA upstream, TPUs that should not feel like foreign add-ons, node auto provisioning and custom compute classes. Janet Kuo describes AI conformance as a superset of regular conformance, created because AI workloads bring their own networking and accelerator requirements. The talk is friendly and technical, with the jargon left in and the jokes kept light. A release-lead episode gives the widest view of how the project moves. The SIG interviews suit anyone who wants to go deep on one area.
Made for: Suits a listener who runs or builds on Kubernetes and wants release changes, deprecations and SIG decisions explained by the maintainers and release leads who made them.
What sets it apart: The show is from Google and features GKE and Google Cloud engineers, but it also covers upstream community work in depth: it has a standing tradition of interviewing each Kubernetes version's release lead, alongside SIG maintainers and the docs team.
A biweekly podcast focused on what's happening in the Kubernetes community hosted by Abdel Sghiouar and Kaslin Fields. We cover Kubernetes, cloud-native applications, and other developments in the ecosystem. Abdel and Kaslin on Twitter at @KubernetesPod or by email at [email protected].
A Postgres pod enters; Abdel Sghiouar asks the question your StatefulSet shortcut would rather dodge.
Press play if you want
Short reviews from the PlayNext desk, based on the episodes we processed.
A grounded Kubernetes 1.35 review that links in-place pod resource updates with the coordination work behind a major open source release.
Kubernetes 1.35 lands here as a release shaped by people as much as by features. Release lead Drew comes through as a practical operator, balancing a chaotic stretch with DevOps work and occasional SRE major-incident duties while keeping the project moving.
A grounded GKE anniversary episode on keeping Kubernetes adaptable for AI without losing its core.
GKE's tenth anniversary gives this Kubernetes Podcast episode a useful reason to look at where the product has been and where Kubernetes is heading as AI workloads put new pressure on infrastructure.
A grounded look at Gateway API, IPv6 limits, and Kubernetes networking work.
Antonio Ojea gives the episode a steady technical center: Kubernetes networking APIs are carrying more weight while many teams still live with Ingress.
Kubernetes documentation feels like release work once SIG Docs' steady role comes into focus.
Kubernetes documentation comes across here as steady release work, not a static library sitting beside the code. Technical writer Shannon Kularotna describes how SIG Docs helps keep the project moving through blog reviews, reference documentation, and coordination with the…
Linux talk with hands-on scars.
Ruby talk with the edges left on.
This episode covers news about Dragonfly, Red Hat Terraform provider for ROSA, Azure Operator Nexus, HashiCorp license change, and S2C2F framework, plus an interview with Leaksignal on data exfiltration prevention in service meshes.
This episode covers news about Notary Project releases, Kubernetes package repository migration, Gateway API service mesh support, and an interview with Phil Estes about containers and ContainerD.
This episode covers news about Linkerd 1.24, Amazon's investment in Anthropic, KubeCon CFP, and an interview with Istio maintainers.
This episode covers Kubernetes news from KubeCon, including OpenTelemetry expansion, Gitpod leaving Kubernetes, OpenCost incubation, and an interview about scaling Kubernetes for AI training.
This episode covers Kubernetes 1.28 release with lead Grace Nguyen and tech news including Docker Desktop updates, CNCF End User TAB, Go 1.21, and ConfigU funding.
Kubernetes Podcast from Google hosts Kaslin Fields and Abdel Sigiwar speak with Drew Hagen, the release lead for Kubernetes 1.35, which ships as Timbernetes.
Kubernetes Podcast from Google opens with community news on Knative graduating at CNCF maturity, LLMD 0.3, the Certified Meshery Contributor credential, and Headlamp for Carpenter.
Antonio Ojea argues that Kubernetes networking is shifting from static network hardware to API-driven primitives like dynamic resource allocation and Gateway API.
Shannon Kularotna describes starting Kubernetes documentation contributions about five years ago without knowing what open source meant, after hearing contributions could help with interviews.
Kubernetes Podcast from Google centers this episode on Kubernetes 1.34 and Vyom Yadav's Of Wind & Will theme.
Kubernetes Podcast from Google interviews Clayton Coleman and Rob Shaw about why large language models require different Kubernetes serving and scheduling primitives than web applications.
Kaslin Fields and Steve McGee interview Ben Good about using Kubernetes and related systems to build platforms for engineering teams.
Pierre-Gilles Mialon and Glenn Yu discuss GitOps, Kubernetes Resource Model infrastructure, policy as code, platform engineering, Nomad, Argo CD, and FluxCD with Kubernetes Podcast hosts Kaslin Fields and Abdel Sigiwar.
Ricardo Rocha, who leads platform infrastructure at CERN, explains why scientific computing needs Kubernetes scheduling features such as queues, quotas, priorities, preemption, and gang scheduling.
Alain Rignier of KuboLabs and Camilla Martins, a Google Cloud Developer Expert and community organizer, discuss KubeCon EU, Google Cloud Next, and cloud-native trends.
Nick Eberts and John Lee explain Multi-Cluster Orchestrator, an open-source tool announced at KubeCon EU 2025 for placing workloads across Kubernetes clusters with scarce GPU and TPU capacity.
Kaslin Fields, Mofi Rahman, and Abdel present live KubeCon EU 2025 conversations recorded in London about platform engineering, Kubernetes components, AI workloads, and a four-node farm cluster.
Kaslin Fields and Abdel Sigiouar interview Kubernetes 1.33 release lead Nina Polshakova about 64 enhancements, compared with 44 in Kubernetes 1.32.
Jesse Butler of AWS and Nick Slattery of Google explain Kubernetes Resource Orchestrator, or KRO, a cloud-agnostic project for grouping Kubernetes and cloud resources behind simpler APIs.
The Kubernetes Podcast from Google interviews LinkedIn software engineers Ahmet Alp Balkan and Ronak Natani about moving stateless, stateful, and batch workloads from a homegrown container and scheduler stack to Kubernetes.
Mophie Raman and Kazlyn Fields interview Lior Lieberman, a Google Cloud software engineer and Gateway API contributor, about migrating from Kubernetes Ingress to Gateway API.
William Morgan, CEO of Buoyant and creator of Linkerd, explains how the service mesh adds HTTP retries, timeouts, load balancing, and mutual TLS to Kubernetes without application changes.
John Belomarek, a Google senior staff software engineer and co-chair of Kubernetes Working Group Device Management, explains why AI workloads require more precise hardware allocation.
Kubernetes Podcast from Google reviews 2024 with Kathleen, Mofi, and Abdel, reporting about 300,000 downloads and a more consistent every-two-weeks publishing schedule.
Frederico Muñoz, release lead for Kubernetes 1.32, explains why the release is themed Penelope and how its weaving metaphor reflects Kubernetes feature lifecycles.
Kubernetes Podcast from Google covers KubeCon North America 2024 through CNCF news and attendee interviews recorded on the show floor.
Maciek Różacki and Wojciech Tyczyński explain why GKE now supports 65,000-node clusters for AI training and inference workloads.
Yuan Tang and Carlos Eduardo Arango Gutierrez describe Kubernetes Working Group Serving, created after KubeCon Europe in Paris to address model-serving and inference challenges.
Michelle Shoubirka, a cloud security developer advocate at Google, argues that asking whether VMs or containers are more secure is incomplete because architecture, trust boundaries, and organizational collaboration determine the result.
Marvin Beckers describes KCP as a cloud-native control plane for Kubernetes-like APIs that removes container orchestration from the Kubernetes resource model.
David Shah and Evan Regmi explain how Spotify's Hendrix ML platform gives internal AI researchers and ML practitioners access to compute without requiring direct Kubernetes or Ray expertise.
Solomon Hykes, co-founder of Dagger and creator of Docker, explains Dagger as an engine for running pipelines and containers across environments.
Mophie Rahman interviews AnyScale’s Richard Liao and Kai-Hsun Chen about Ray, an open-source compute engine for scaling AI workloads, and KubeRay, its Kubernetes operator.
Karthik Satchitanand, principal software engineer at Harness and Litmus co-founder, explains chaos engineering as controlled failure injection tested against a "steady state hypothesis." He contrasts game days and one-off disaster recovery tests with continuous resilience, where experiments green-light every deployment.
The Kubernetes Podcast from Google examines Kubernetes 1.31, “Elli,” with release lead Angelos Kolaitis, a senior software engineer at Canonical.
Ohad Maislish, CEO and co-founder of EnvZero and an OpenTofu co-founder, explains why HashiCorp’s August 2023 license change led to a Terraform fork under the Linux Foundation.
Mauricio Salatino explains Dapr as a distributed application runtime that exposes APIs for state, messaging, secrets, service discovery, and other infrastructure needs.
Charity Majors, co-founder and CTO of Honeycomb.io, argues that engineering management is a support function connecting team missions to business meaning.
Maciek Sulik explains how Kubernetes batch-controller improvements remove completed pods earlier, preventing large jobs from retaining thousands of pods for days or weeks.
Kubernetes Podcast from Google’s “Leading Kubernetes into its Second Decade” features Nikita Raghunath, Nabarun Pal, and Pako Thu on contributor growth and project leadership.
The Kubernetes Podcast from Google marks Kubernetes’s June 6, 2024 10th anniversary with Tim Hockin and Kelsey Hightower, two contributors involved near the project’s beginning.
Kubernetes Podcast from Google marks Kubernetes’s 10-year anniversary by examining “A Decade of Kubernetes Contribution” with long-time contributors David Eades and Davanam Srinivas.
The Kubernetes Podcast from Google interviews Alvaro Hernandez, founder and CEO of Ongress, about Stackgress, a Postgres distribution designed to run on Kubernetes.
Thomas Pognon and Todd Burt explain how feature flags let teams change application behavior, target beta users, and roll out releases without redeploying code.
Kat Cosgrove, release lead for Kubernetes 1.30, describes a release with 11 alpha, 18 beta, and 17 stable tracked changes after code freeze.
Kubernetes Podcast from Google reports from KubeCon Europe 2024 in Paris, which drew more than 12,000 in-person attendees.
This episode covers Kubernetes and cloud-native news, and features an interview with Matt Klain about Envoy and his new projects.
This episode covers Kubernetes 1.29 release interview, security audit of Caverno project, and Google's Gemini AI model announcement.