Polyaxon v3 is coming →

The control plane for AI engineering.

Polyaxon helps teams build, run, and operate machine learning, AI training, inference, and agent workloads on Kubernetes.

It brings scheduling, experiment tracking, registries, automation, observability, and governance into one workflow. Connect Polyaxon Cloud to your clusters or self-host Polyaxon on infrastructure you control.

  • Open at the coreAn Apache 2.0 core that teams can inspect, extend, and self-host.
  • Infrastructure flexibilityUse Polyaxon Cloud with your clusters or run the control plane yourself.
  • One connected workflowKeep execution, metadata, artifacts, and governance connected.

What we do

One platform across the AI engineering workflow.

Build

  • Notebooks, coding sessions, and sandboxes

  • Experiments, metrics, prompts, code, and artifacts

Run

  • Training, services, pipelines, and agent workloads

  • Queues, presets, and infrastructure connections

Operate

  • Runs, traces, registries, and automation

  • Team access, governance, and operational controls

Built in the open

Polyaxon is built on an Apache 2.0 open-source core. Explore the source, deploy the Community Edition, and join the discussion.