Adrian Moisey is a Staff Site Reliability Engineer (SRE), a Technical Lead for Kubernetes SIG Autoscaling, and a contributor to SIG Network. Jakub Pawełczak is a Cloud Product Manager on Google Kubernetes Engine (GKE) at Google Cloud focused on workload autoscaling.
For years, updating container resources meant paying a steep "restart tax" by killing and rescheduling Pods just to adjust CPU or memory limits. In this episode, we unpack how In-Place Pod Resizing (IPPR) and Startup CPU Boost are changing that, how the Vertical Pod Autoscaler (VPA) integrates with in-place updates, and where Kubernetes autoscaling is heading next.
Do you have something cool to share? Some questions? Let us know:
web: kubernetespodcast.com
mail: kubernetespodcast@google.com
twitter: @kubernetespod
bluesky: @kubernetespodcast.com
News of the week CNCF announces Karmada graduation (and Karmada v1.19 release)
llm-d reaches version 1.0 under CNCF governance
GKE Public Preview: Combining Vertical Pod Autoscaler (VPA) with Horizontal Pod Autoscaler (HPA)
GKE Standard: In-Place Node Upgrades for CPU node pools
GKE Sandbox: General Availability of the microVM sandbox type
KubeCon + CloudNativeCon North America 2026
CNCF October Update: Passing 500 Golden Kubestronauts & .project metadata governance
Kubernetes v1.37: KubeletInUserNamespace (aka Rootless mode) Graduates to Beta (and Kubernetes Podcast Episode 271: Kubernetes 1.37, with Dipesh Rawat)
Kubernetes v1.34 reaches End of Life
Volcano v1.16 released
WSL Containers (WSLc) now Generally Available on Windows 11 (Cloud Native Now coverage)
Links from the interview KEP-1287: In-Place Update of Pod Resources (IPPR)
Resize CPU and Memory Resources assigned to Containers (Kubernetes Docs)
Kubernetes Autoscaler & Vertical Pod Autoscaler (VPA) on GitHub
Accelerate application startup with CPU startup boost in GKE
Google Cloud Blog: GKE CPU startup boost: Faster pod starts, lower costs
JAVAPRO Magazine (09-2026): The Next Chapter of Kubernetes and Java: Java Apps and Dynamic Resource Allocation (Direct PDF)
Kubernetes SIG Autoscaling & Kubernetes Slack (#sig-autoscaling / #vertical-pod-autoscaler)
Kubernetes SIG Contributor Experience
Specifying a Disruption Budget for your Application (PDB)
Karpenter & KEDA
KubeRay (Ray on Kubernetes)
Kubernetes Podcast Episode 272: Agent Substrate, with Tim Hockin and Brandon Royal
At 100,000-accelerator scale, something fails multiple times an hour, which is why Google's Chief Technologist for AI Infrastructure Amin Vahdat thinks FLOPS is a vanity metric. The metric that matters is what he calls goodput: the useful work a workload actually delivers through real failures. Amin walks through the calculus that split the TPU line into 8i and 8t for the first time, why the TPU's core primitives haven't changed since v1, and how Google and DeepMind co-design in the same rooms, intercepting chip architectures mid-flight before tape-out. He explains why long-horizon agents are sending demand for CPUs and storage through the roof alongside accelerators, how optical circuit switches reroute light to a spare rack in milliseconds, and why Google would rather wait on a utility than build its own gigawatt. We also cover orbital data centers and the multi-megawatt rack of 2036.
Hosted by Sonya Huang, Sequoia Capital
0:00 – Introduction
1:47 – What makes a data center an AI data center
5:30 – Goodput, not FLOPS: holding yourself accountable when something fails every hour
11:52 – Doubling token capacity every six months, and where the gains actually come from
15:32 – The TPU bet: from a contrarian call in 2013 to splitting 8i and 8t
23:30 – The case for and against co-design
26:11 – Shoulder to shoulder with DeepMind: intercepting chips mid-flight
34:16 – Long-horizon agents change the shape of the data center
37:50 – Optical circuit switching and the state of networking
42:35 – Power is the binding constraint: utilities, gigawatts, and sizing a data center
49:23 – Training vs. serving clusters, seven-year-old TPUs, and open standards
58:29 – Orbital data centers and the supercomputer of 2036
In ICYMI Episode #1024, Jon Krohn tracks the gap between AI investment and AI return, from the technology side to the people side. Hear from Pete Johnson, Jerry Yurchisin, Priyanka Vergadia and Tristan Handy, discussing why four out of five organizations have the structures for AI success in place while only one in five sees the returns, which decisions should never be handed to a language model however confident it sounds, how to structure Claude skills so that your output stops being slop and why the semantic layer matters more, not less, now that analytics agents are the ones asking the questions.
Additional materials: www.superdatascience.com/1024
Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.
In this episode you will learn:
(00:56) Vector Search, Agentic Memory and Effective RAG
(09:20) Mathematical Optimization in the Agentic AI Era
(17:30) Anyone Can Write Code Now, So What Gets You Hired?
(27:14) How dbt Won Analytics Engineering
Tim Hockin is a long term software engineer with Google Cloud and I would argue one of the fathers of Kubernetes. Brandon Royal is a product manager on GKE and has been behind the launch of multiple OSS projects like the Ray Operator for k8s, Agent Sandbox and our topic for today - Agent Substrate.
Do you have something cool to share? Some questions? Let us know:
- web: kubernetespodcast.com
- mail: kubernetespodcast@google.com
- twitter: @kubernetespod
- bluesky: @kubernetespodcast.com
News of the week Kubernetes 1.37: Garhwal
Kubernetes Podcast: Kubernetes 1.37 with Dipesh Rawat
Kubernetes release & maintenance information: https://kubernetes.io/releases/
BOMHort
Google Open Source Blog: Adapting open source practices to an AI-first world: A retrospective on 2025
Kubernetes-security-announce Google Group
Kubernetes CVE RSS Feed
Links from the interview Agent Substrate on GitHub
Agent Sandbox
Kubernetes Checkpoint & Restore Working Group
spiffe.io
HashiCorp Blog: SPIFFE: Securing the identity of agentic AI and non-human actors
Links from the post-interview chat Kubernetes Podcast: Agent Sandbox and Lovable, with Jonathan Grahl
In Episode #1019, Priyanka Vergadia (founder of The Cloud Girl, former Senior Director of AI Transformation at Microsoft and Head of North America Developer Relations at Google) joins Jon Krohn to explain why almost every company has bought AI tools and almost none of them are seeing a return. Her fix is a budget split that will make any CFO wince: seven dollars on training employees for every dollar spent on the tools themselves. Having spent a decade turning dense cloud and AI concepts into sketches that a quarter-million developers actually remember, and having carried GitHub Copilot into Fortune 100 boardrooms, she has watched the gap between tool purchase and real production use up close. In this episode, Priyanka defines the elusive quality she calls taste, walks through how she structures Claude skills so her output stops being slop, unpacks her 10-20-70 framework, and shares breaking news about what she is building next.
Additional materials: https://www.superdatascience.com/1019
Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.
In this episode you will learn:
(00:10:39) What “taste” actually means and why Priyanka now interviews for it
(00:31:39) How to build a Claude skill by breaking a task into explicit sub-tasks
(00:36:11) The 10-20-70 framework for AI budgets
(00:47:52) The weekend exercise for finding what makes you different
Brought to You By:
• Antithesis – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages.
• Buildkite – CI software built to absorb whatever your coding agents throw at the build queue
• Sentry – application monitoring software considered “not bad” by millions of developers
—
Kelsey Hightower went from a self-taught technician installing DSL modems to becoming one of Google’s elite Distinguished Engineers, whom the CEO of Microsoft personally tried to recruit. Hightower’s career achievements are rooted in hard work and self-directed learning, and today he’s one of the most influential voices in modern infrastructure, through his talks, open source work, and writing.
In this episode of The Pragmatic Engineer podcast, Kelsey and I cover his unconventional path into tech and the lessons he’s learned during three decades in the industry. We discuss his entrepreneurial years, building a reputation through open source, the rise of containers and Kubernetes, and his time at Google during one of the most consequential periods in cloud computing.
He recounts how a job offer from a big tech giant led to the biggest raise of his career, what prompted him to slow down after years of career acceleration, and we also discuss his perspective on AI. Throughout, Kelsey keeps a simple idea front of mind: that technology is ultimately about people. Whether it’s infrastructure, leadership, careers, or AI, he argues that the goal is not to build technology for its own sake; it’s to solve meaningful human problems.
—
Timestamps
00:00 Intro
03:34 Kelsey’s first job at McDonald’s
05:04 His non-traditional path into tech
11:45 Landing his first tech job with an A+ certification
15:33 His entrepreneurial years
19:45 Joining Google as a data center technician
27:48 Learning automation at a Rackspace spinoff
33:26 Moving into financial services
50:00 Building a reputation through open source
53:55 From configuration management to containers
1:08:20 The rise of Kubernetes
1:25:05 Why he almost joined NASA instead of Google
1:29:20 Defining DevRel at Google
1:38:20 Demonstrating impact at Google
1:41:20 Microsoft's offer
1:55:20 Learning how to slow down
2:06:39 Advising and investing
2:15:03 A people-first view of GenAI
2:24:27 Using AI with guardrails
2:28:26 Matching AI to the task
2:36:06 Staying relevant in the AI era
—
The Pragmatic Engineer deepdives relevant for this episode:
• Career paths for software engineers at large tech companies
• The past and future of modern backend practices
• How Kubernetes is built
• How Linux is built
• The Staff Engineer’s Path: You’re a role model now (sorry!)
—
Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email podcast@pragmaticengineer.com.
Get full access to The Pragmatic Engineer at newsletter.pragmaticengineer.com/subscribe
At Google Cloud Next 2026, Finout co-founder and CEO Roi Ravhon and Google Cloud FinOps lead Pathik Sharma discussed how FinOps is rapidly evolving for the AI era. Ravhon argued that while cloud FinOps had a decade to mature, AI economics are forcing the industry to adapt within a year. Unlike traditional cloud workloads, AI costs are unpredictable because token usage varies even for identical prompts, while advanced reasoning models consume significantly more tokens despite falling prices.
Both emphasized that effective AI FinOps requires intelligent orchestration, routing workloads to the cheapest capable models instead of defaulting to expensive frontier models. Sharma noted that AI costs extend beyond APIs to GPUs, storage, training, and organizational adoption. They also cautioned against relying solely on LLMs for operational automation. Deterministic systems, observability metrics, and human approvals remain essential guardrails. Ultimately, both stressed that FinOps is primarily an organizational and cultural discipline, recommending newcomers start with the FinOps Foundation before investing in tools.
Learn more from The New Stack around the latest in FinOps:
Why FinOps Isn’t About Saving Money
FinOps Foundation’s FOCUS 1.2 Expands to SaaS, PaaS
Join our community of newsletter subscribers to stay on top of the news and at the top of your game.
At KubeCon Europe, Google Cloud’s Jago Macleod and Abdel Sghiouar argued that adopting Arm for Kubernetes workloads has shifted from a complex migration to a practical, low-friction choice. After a year of production use, Google’s custom Arm-based Axion processors—powering C4A and N4A instances—are positioned as broadly viable for most containerized applications, offering strong gains in performance, cost efficiency, and energy usage compared to x86.
Rather than requiring a full overhaul, moving to Arm typically involves recompiling containers for a multi-architecture target and gradually rolling out via Kubernetes practices like canary deployments. While edge cases exist, they are relatively uncommon.
A key enabler is GKE’s compute classes, which allow workloads to express preferences across VM types, turning infrastructure decisions into automated scheduling choices rather than manual provisioning.
Ultimately, the conversation points to a larger constraint: energy. As AI workloads grow, efficiency—measured in “tokens per watt”—is emerging as the defining metric, with cost savings translating directly into greater compute capacity.
Learn more from The New Stack about the latest developments around Google’s work with Axion:
Arm: See a Demo About Migrating a x86-Based App to ARM64
Do All Your AI Workloads Actually Require Expensive GPUs?
Join our community of newsletter subscribers to stay on top of the news and at the top of your game.
Startup founders are being pushed to move faster than ever, using AI while facing tighter funding, rising infrastructure costs, and more pressure to show real traction early. Cloud credits, access to GPUs, and foundation models have made it easier to get started, but those early infrastructure choices can have unforeseen consequences once startups move beyond free credits and into real cloud bills.
On this episode of TechCrunch's Equity podcast, Rebecca Bellan caught up with Darren Mowry, Google Cloud’s vice president of global startups who is right at the center of those tradeoffs. Together, they discuss what Mowry’s seeing across the startup ecosystem, how Google Cloud is competing for AI startups, and what founders should be thinking about as they scale.
Listen to the full episode to hear about:
How Google positions against AWS and Microsoft in the AI startup race.
TPUs vs GPUs: How much does hardware choice matter for early-stage companies?
Which AI verticals are seeing real growth, and what’s standing out in biotech, climate tech, developer tools, and world models.
What red flags will signal that a startup isn’t going to make it.
Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify and all the casts. You also can follow Equity on X and Threads, at @EquityPod.
Learn more about your ad choices. Visit megaphone.fm/adchoices
AI is good at seeing patterns, but it’s humans who figure out what to do next, says technologist Priyanka Vergadia. She shares three stories of human excellence sparked by AI insights and offers a pathway to identify and cultivate your irreplaceable qualities, turning the AI revolution from a threat into an opportunity.
Hosted on Acast. See acast.com/privacy for more information.
Ryan sits down with Dan Ciruli, VP and General Manager of Cloud Native at Nutanix, to talk about getting your virtual machines and Kubernetes to play nice in cloud-native environments, why VMs are still relevant in enterprise applications, and how AI can help modernize legacy systems.
Episode notes:
Nutanix combines compute, storage, virtualization, and networking so you can run applications and manage data across on-premises datacenters, public clouds, and edge locations all on one platform.
Connect with Dan on Linkedin and Bluesky.
Congrats to Necromancer badge winner David Ferenczy Rogožan! They won the badge on their answer to Where does adb shell mkdir create directories.
TRANSCRIPT
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
GPUs dominate today’s AI landscape, but Google argues they are not necessary for every workload. As AI adoption has grown, customers have increasingly demanded compute options that deliver high performance with lower cost and power consumption. Drawing on its long history of custom silicon, Google introduced Axion CPUs in 2024 to meet needs for massive scale, flexibility, and general-purpose computing alongside AI workloads. The Axion-based C4A instance is generally available, while the newer N4A virtual machines promise up to 2x price performance.
In this episode, Andrei Gueletii, a technical solutions consultant for Google Cloud joined Gari Singh, a product manager for Google Kubernetes Engine (GKE), and Pranay Bakre, a principal solutions engineer at Arm for this episode, recorded at KubeCon + CloudNativeCon North America, in Atlanta. Built on Arm Neoverse V2 cores, Axion processors emphasize energy efficiency and customization, including flexible machine shapes that let users tailor memory and CPU resources. These features are particularly valuable for platform engineering teams, which must optimize centralized infrastructure for cost, FinOps goals, and price performance as they scale.
Importantly, many AI tasks—such as inference for smaller models or batch-oriented jobs—do not require GPUs. CPUs can be more efficient when GPU memory is underutilized or latency demands are low. By decoupling workloads and choosing the right compute for each task, organizations can significantly reduce AI compute costs.
Learn more from The New Stack about the Axion-based C4A:
Beyond Speed: Why Your Next App Must Be Multi-Architecture
Arm: See a Demo About Migrating a x86-Based App to ARM64
Join our community of newsletter subscribers to stay on top of the news and at the top of your game.