Skip to main content

Dev Station Technology

AI Development

Kubeflow development services for teams that run models on Kubernetes

We have no published Kubeflow build. We have shipped a Docker multi-stage deployment and we have no published Kubernetes build, so we say plainly that a Kubeflow job starts with a cluster.

0Published builds on Kubeflow
26.03Latest Community Distribution listed
CNCFGraduated project, per its releases page
0Published Kubernetes builds
What we build

What we would take on with Kubeflow

Pipelines on a cluster

Training and serving steps run as jobs on Kubernetes.

Container images

Multi-stage Docker builds, as on TruckFleet.

Tracking beside it

Runs and results recorded, with MLflow where it fits.

Distribution upgrades

Moves onto the current Community Distribution.

Architecture

Where Kubeflow would sit

Kubeflow on top of a cluster
Images
Docker logo
Docker multi-stage buildTruckFleet Labs
Cluster
Kubernetes logo
KubernetesNot yet a published build
ML platform
Kubeflow logo
KubeflowWhere it would sit
MLflow logo
MLflowTracking, if used

Illustration. A dashed outline marks where the technology would sit. We have no published build on it.

Releases

What the Kubeflow releases page shows

Kubeflow on its releases page
Community Distribution 26.03Listed first, with master and v26.03 in the version switcher.
CNCF graduatedA banner says Kubeflow is now a CNCF graduated project.
Page modified 13 June 2026The page shows its last modified date, and no release dates or cadence.

Source: Kubeflow documentation, Releases, kubeflow.org/docs/releases, checked 6 November 2026.

Builds

The closest build we can show you

TruckFleet Labs dispatch console and driver app built by Dev Station
Logistics, fleet operations

TruckFleet Labs

A FastAPI back end deployed with a Docker multi-stage build and Docker Compose.

38Drivers on the app by week three
Fit check

When Kubeflow is the right call

Pick Kubeflow when

  • You already run Kubernetes
  • Several people train and serve models
  • Pipelines must be repeatable

Look elsewhere when

  • You need a published Kubeflow build to inspect, which we do not have
  • One model on one host, where Docker Compose is enough
  • You have no cluster and no one to run one
Stack

What Kubeflow usually sits beside

Kubernetes logoKubernetes
MLflow logoMLflow
Apache Airflow logoApache Airflow
Docker logoDocker
PyTorch logoPyTorch
TensorFlow logoTensorFlow
FAQs

What buyers ask first

Which Kubeflow release is current?

The releases page lists Kubeflow Community Distribution 26.03 first. It gives no release date and no cadence.

Do you have a published Kubeflow build?

No. We say so on the first call and show the nearest work, which is a Docker multi-stage deployment, with no published Kubernetes build.

Is Kubeflow a CNCF project?

A banner on its releases page says Kubeflow is now a CNCF graduated project.

Who owns the code and the cloud accounts?

You do, from the first commit. Repositories and cloud accounts stay in your name.

Tell us how you train and serve models

Thirty minutes on your cluster, your pipelines and your models. No sales pitch.

Let's Talk