Radio Cloud Native by Mirantis

Hosted by Mirantis Inc.

Technology News

Website RSS feed LinkedIn X / Twitter Facebook YouTube

Episodes

Latest episode

Nov 2026

Language

EN-US

About the show

Welcome to Radio Cloud Native by Mirantis, your go-to podcast for all things cloud native and Kubernetes. Join us as we explore the latest trends, innovative solutions, and open source projects shaping the modern cloud landscape. From enterprise solutions to open source integrations, we delve into everything you need to know to stay ahead in the world of cloud native and Kubernetes. For more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com Learn more about Mirantis at https://www.mirantis.com/

Listen to episodes

60 recent

November 8, 2026Episode 7047 min

The Hidden Gaps in Agentic AI: Trust, Evaluation, and Security

NOTE : this podcast was recorded on 7/14/2026. In this installment of Radio Cloud Native, two engineers from the Office of the CTO at Mirantis: Justin Winter (Principal Software Engineer, AI/Agentic Systems) and Randy Bias (VP of Strategy and Technology, Open Source Program Office), sit down for an unscripted conversation about what it actually takes to build with AI agents in 2026. The conversation moves from the personal to the practical: how each of them got pulled into AI early (one as a builder, one as a skeptic-turned-convert), and what's changed as they've gone from experimenting freely to shipping production systems that others depend on. Along the way, they get into the unglamorous but critical problems nobody's fully solved yet: how do you actually trust what an agent produces, how do you evaluate a system that's non-deterministic by design, and how does security keep up when the tools are moving faster than the policies meant to govern them? They close with a shared take on where all of this sits in the bigger arc of technological change, and what they'd tell someone just getting started. Topics include: Why coding harnesses like Claude Code, Codex, and Fable matter more right now than the underlying models themselves The evals gap most teams haven't solved yet and why "black-boxing" your agents is a problem How Agent Skills quietly became the pattern that stuck, while MCP is "still finding its spot" The case for small, cheap local models handling routine workflow steps while frontier models handle judgment calls Why AI security teams are structurally behind the curve, plus a live example of Claude flagging a harmless dev secret as a critical risk What separates people who get real value from AI tools from those who don't (hint: it's not prompting skill) Why we're still in the "2011–2012 of cloud computing" phase of this AI shift, with several more years of rapid change ahead If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

October 11, 2026Episode 7134 min

How eBPF and Cilium Are Reinventing Kubernetes Networking

Why doesn't traditional network security work in Kubernetes anymore - and what's replacing it? In this episode of Radio Cloud Native, Nick Chase of Mirantis sits down with Christian Hernandez and Nico Vibert of Isovalent (Cisco) to unpack how eBPF and Cilium are rewriting the rules of cloud native networking. They cover why IP-based security breaks down in dynamic Kubernetes environments, how identity-aware policies (labels and metadata) are taking its place, and what eBPF actually does at the kernel level to make tools like Cilium, Tetragon, and Hubble possible. They also dive into Cilium's place as an ideal CNI for Kubernetes for policy driven networking, as well as a new component they are dubbing the "Kubernetworker" (combining Kubernetes and Networking). The group will finish looking towards the future of networking and the changes that are still on the horizon as AI workloads become commonplace. A full list of topics covered in this episode includes: Current networking landscape What is eBPF? Traditional vs K8s networking K8s networking training/certification What networking changes mean for Cilium Mirantis k0rdent + Cilium partnership Future of networking (for AI/ML) If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

May 31, 2026Episode 6918 min

All Infrastructure Is AI Infrastructure: Navigating the New Era of Enterprise AI at Scale

As enterprises scale generative AI and autonomous agents, every layer of the technology stack - from silicon to orchestration (ie. Metal to Model) - is being reshaped. In this episode, hosted by Mirantis CMO Dominic Wilde, you will hear why AI workloads reverse decades of cloud abstraction, what full-stack AI infrastructure actually demands (GPUs, CPUs, networking fabrics, power, storage, etc.), how data sovereignty and regulatory frameworks like GDPR and DORA create hard architectural requirements, and what principles - manageability, observability, flexibility, repeatability, and the importance of open infrastructure - must be used to guide enterprise AI platforms into the future. All of this change has informed how the teams at Mirantis built k0rdent, a complete, open source platform for AI infrastructure orchestration. A full list of topics covered in this episode include: The end of abstraction and what's next Full-stack AI infrastructure Data/AI sovereignty & compliance The developer experience gap Core principles for enterprise-ready AI platforms The case for open infrastructure Commoditization to contracts k0rdent by Mirantis If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

May 10, 2026Episode 6829 min

Rancher Kubernetes Engine (RKE1) is EOL. So What's Next?

Rancher Kubernetes Engine 1 (RKE1) reached end of life on July 31st of 2025, meaning no more security patches, bug fixes, or a direct upgrade path forward. For businesses still running RKE1, the clock is ticking - and the decisions made now will have major implications for security, compliance, and operational continuity. In this episode, Edward Ionel of Mirantis is back to walk you through what re-platforming actually looks like in practice, the hidden costs and challenges of migration, and how to evaluate your options for moving to RKE2, RKE3, or a more modern Kubernetes platform. Whether you're managing 10 clusters or hundreds, this is a practical guide to navigating one of the most consequential infrastructure transitions your team may face. A full list of this episode's topics include: Official announcement of EOL for RKE1 Why migrating away from RKE1 doesn't work Current evolution in App infrastructure Current evolution of Cloud Common containerization challenges k0rdent by Mirantis Full container solution stack Value of the cloud native/K8s-first approach If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

April 12, 2026Episode 6721 min

Stop Losing Money on Idle GPUs

The GPU cloud market is booming - but for NeoClouds and cloud service providers, explosive growth is quickly becoming a race to the bottom. In this episode, Edward Ionel of Mirantis breaks down why raw compute is no longer enough to sustain a profitable business, and what operators must do to protect their margins before the window closes. Edward walks through the full monetization maturity model - from commoditized bare metal to a fully managed AI factory platforms capable of delivering 50%+ profit margins - and explains why evolving from selling compute to selling outcomes is no longer optional. If you're building or operating GPU infrastructure, this episode is a must-listen. Reach out to Edward directly for a demo: eionel@mirantis.com A full list of this episode's topics include: Emerging GPU Cloud and AI Infrastructure opportunity Evolution and challenges for NeoClouds Transforming AI Services GPU depreciation gap and market erosion The 4 stages of GPU monetization Building an AI Service platform Building an AI Factory A hyperscaler-like experience for AI Real world case study If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

March 8, 2026Episode 6611 min

A Quick Introduction to gRPC and Related Tools

In this month's episode Sr. Solution Architect at Mirantis, Zachary Fanelle, will walk you through the basics for getting started with gRPC and how to easily start integrating it into your Kubernetes workflows. He will cover HTTP and REST APIs, as well as how these web technologies prompted Google to create gRPC in the first place. You will gain practical knowledge about what gPRC is, why it is leveraged by many developers in many use cases, and how it is supported in Kubernetes as of K8s version 1.24. A full list of this episode's topics include: What is gRPC? Why leverage gRPC? gRPC use cases Drawbacks of gRPC How to get started with gRPC If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

February 8, 2026Episode 6524 min

Defeat Platform Complexity with Use Case-Optimized Multi-Cluster Kubernetes

In this month's installment, Sergey Goncharov, Sr. Product Manager for Mirantis Kubernetes Engine 4k (MKE 4k) is back to help you tackle the ever-emerging problem of Platform Engineering for complex multi-cluster, multi-cloud Kubernetes workloads. With the rise of AI/Ml, these are becoming more and more prevalent in enterprises around the world. Overcoming this complexity requires a multi-cluster approach: designing clusters tailored to specific workloads with declarative services and policies, ensuring consistent lifecycle management (LCM), and scaling this strategy across your environments. By optimizing platforms for specific use cases (Kubernetes-based platforms that is), you can improve agility, streamline operations, boost workload availability, enhance security, and maintain compliance - without implementing a global multi-cluster architecture at smaller scales. A full list of this episode's topics include: Challenges with Kubernetes complexity Multi-cluster thinking is key Building use case-optimzied platforms at scale Example open source tools for multi-cluster K8s If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

January 11, 2026Episode 6424 min

Cloud Native Virtualization Best Practices for Public Sector and Highly Regulated Clouds

In this episode of Radio Cloud Native, Michael Tillman, Sr. Solutions Architect and FedGov Cloud expert, is back to provide you with an in-depth overview of everything you need to know about Public Sector cloud native virtualization and modernization. Michael will cover the challenges most FedGov cloud customers face in terms of Legacy system modernization before presenting Mirantis Kubernetes Engine 4k (MKE 4k) as the ideal solution for your FedGov cloud needs. Furthermore, Michael will provide real world use case examples (anonymized to maintain the privacy of our customer) showcasing ways in which Mirantis was able to drastically improve the digital transformation process for its Public Sector clientele - all while maintaining the reliability and security needed in the most highly regulate industries. A full list of this episode's topics include: Legacy Public Sector IT challenges Cloud Native Virtualization Mirantis Kubernetes Engine 4k (MKE 4k + KubeVirt) Public Sector IT best practices Conclusion and summary If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

December 7, 2025Episode 6328 min

The New Mirantis Kubernetes Engine (MKE) with KubeVirt Integration

In this month's installment, Sr. Product Manager for Mirantis Kubernetes Engine (MKE), Sergey Goncharov, will walk you through how to achieve full cloud native virtualization on top of Kubernetes-based infrastructure using MKE 4k (our latest release) and open source KubeVirt. By the end of this talk, you'll be able to run Kubernetes and/or Swarm-based VMs alongside containers and even within the K8s clusters themselves! A full list of topics in this episode include: What are virtualized workloads & what do they do? Rising trend of Kubernetes-based Infrastructure Why migrate to Kubernetes? How to migrate to Kubernetes? Leveraging Open Source KubeVirt The Mirantis solution: Mirantis Kubernetes Engine 4k (MKE 4k) If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com

November 2, 2025Episode 6239 min

OpenStack Storage Explained: From Hardware to Strategy

In this month's installment, we're diving into OpenStack storage: the challenges, the solutions, and the best practices for securely and efficiently storing your bits when leveraging an OpenStack cloud. Proctored by Christian Huebner, Director of Services Architecture at Mirantis, this episode will build off his 25+ years of experience with cloud storage, specifically in the context of OpenStack. This episode is perfect for OpenStack admins that are looking for better ways to deal with storage, as well as anyone working with OpenStack clouds. A full list of topics for this episode include: Computer storage background (general) Types of storage (OpenStack) OpenStack storage backends (Ceph) Software Defined Storage (SDS) alternatives Storage Area Networks (SANs) Shared files systems (NFS) Local storage (Flash) Combining OpenStack storage types Common failure scenarios OpenStack storage security If you want to listen to more episodes of Radio Cloud Native, please visit https://www.mirantis.com/radiocloudnative/ to download, or find them wherever you prefer to consume your podcasts. If you are interested in contributing to Radio Cloud Native, please reach out to our podcast team: podcasts@mirantis.com