deployKF builds machine learning platforms on Kubernetes.
We combine the best of Kubeflow, Airflow†, and MLflow† into a complete platform that is easy to deploy and maintain.
deployKF combines the ease of a managed service with the flexibility of a self-hosted solution.
Our goal is that any Kubernetes user can build a machine learning platform for their organization, without needing specialized MLOps knowledge, or a team of experts to maintain it.
The key features of deployKF are:
- Run on any Kubernetes cluster, including on-premises and in the cloud
- Intuitive centralized configs for all aspects of the platform
- Seamless in-place upgrades and config updates
- Connect your existing Istio, cert-manager, Kyverno, S3, and MySQL
- Use any identity provider via OpenID Connect or LDAP
- Native support for GitOps with ArgoCD
We are always excited to see how and where deployKF is being used!
Here are some stories of deployKF being used in the wild:
Organization Article / Video Cloudflare A look inside the Cloudflare ML Ops platform Have a story to share? Let us know!
To help you get started with deployKF, we have prepared a number of guides:
- ⭐ Getting Started - learn how to run deployKF anywhere
- Local Quickstart - try deployKF on your local machine
- Migrate from Kubeflow Distributions - how and why to migrate from other Kubeflow distributions
For more information about our releases, please see:
deployKF is a new and growing project. If you like what we are doing, please help others discover us by sharing the project with your colleagues and/or the wider community.
We greatly appreciate GitHub Stars ⭐ on the
deployKF/deployKF
repository:
To discuss commercial support options for deployKF, please connect with Aranui Solutions, the company started by the creators of deployKF. Learn more on the Aranui Solutions Website.
The deployKF community uses the Kubeflow Slack for informal discussions among users and contributors.
Please see our community page for more information.
deployKF was originally created and is maintained by Mathew Wicks (GitHub: @thesuperzapper), a Kubeflow lead and maintainer of the popular Apache Airflow Helm Chart. deployKF is a community-led project that welcomes contributions from anyone who wants to help.