Mirantis has announced the open sourcing of k0rdent, a new project that aims to address the complexities of Kubernetes and AI sprawl. In this episode, Randy Bias, VP of Open Source Strategy and Technology at Mirantis, discusses how k0rdent tackles these challenges, its role in the Mirantis open source ecosystem, its impact on platform engineers and developers, and the company’s broader open source strategy.

Managing clusters at scale has become increasingly complex as organizations deploy Kubernetes across multiple infrastructures. The rise of AI-powered applications has further exacerbated this challenge, creating an AI sprawl alongside Kubernetes sprawl. “We can’t manage hundreds of different clusters individually. We need one central location where all those clusters can be connected,” Bias says. k0rdent aims to solve these complexities by providing a single platform for managing clusters. Platform engineers can create, delete, and monitor workloads efficiently while ensuring governance and disaster recovery capabilities.

While k0rdent is primarily targeted at platform engineers, Bias explains that it indirectly benefits developers by streamlining the deployment process. Despite its extensibility, Kubernetes presents challenges for developers due to its complexity. k0rdent abstracts much of this complexity by enabling platform engineers to create “golden paths”, standardized workflows that help developers deploy applications more efficiently without being burdened by Kubernetes’s intricacies.

Within Mirantis’s open source ecosystem, k0rdent acts as a super control plane. Bias describes how it builds upon K0s, Mirantis’s lightweight Kubernetes distribution, and k0smotron, its distributed control plane. While k0rdent uses these by default, it can be integrated with other tools, making it a versatile solution for managing multi-cluster environments. k0rdent also provides observability, telemetry, and financial operations (FinOps) capabilities, to ensure better cost management and operational visibility.

Bias discusses why Mirantis’s commitment to open source is central to k0rdent’s development and serves as a foundation for future innovations. The project is designed to be Kubernetes-native, leveraging Cluster API and other open source components to provide an unbiased, extensible platform. Bias explains that Mirantis’ goal is to create a multi-cluster, multi-cloud fabric that enterprises can use to build and manage their internal developer platforms (IDP) without vendor lock-in.

Mirantis also plans to introduce a commercial version of k0rdent, k0rdent Enterprise, which will boast additional features such as secure supply chain management, validated service templates, and enterprise support for platforms like vSphere and Active Directory (AD). Bias confirms that while the open source project will remain freely available, the enterprise version will cater to businesses needing enhanced security, compliance, and support.

The discussion shifts to the broader open source AI landscape, particularly in response to the emergence of DeepSeek. Bias argues that there are no true “moats” in AI, as enterprises generate vast amounts of proprietary data that will fuel open source AI models. Bias envisions AI adoption accelerating within enterprises, further intensifying the need for k0rdent to manage AI-powered workloads at scale. As AI becomes an integral part of modern applications, Bias believes that enterprises will require robust solutions to orchestrate and govern their expanding AI infrastructures.

Guest: Randy Bias (LinkedIn)
Company: Mirantis (Twitter)
Show: Open Source Means Business

This summary was written by Emily Nicholls.