Runners
Runners allows Kubiya to orchestrate agents
Last updated
Runners allows Kubiya to orchestrate agents
Last updated
A Local Runner is a lightweight Kubernetes operator that enables your AI Teammates and all of their dependencies to run from within your Kubernetes cluster.
Kubiya can use this operator to communicate with your cluster and take care of the heavy lifting of managing the agent's lifecycle - that includes:
Creating deployments and services on your cluster based on agents requirements
Getting agents logs, metrics, and insights
Triggering agents interactions
Hosted runners are Kubiya-hosted runners that have all of the necessary requirements in order to orchestrate your AI Teammate workflows.
Hosted runners can be used for all Kubiya use cases, except for those that involve Kubernetes – those require a local runner.
The main difference between local and hosted runners is that local runners run from within your cluster, whereas hosted runners are in a Kubiya cluster.
You may need a local runner if you want to set up a use case involving Kubernetes or if your organization requires it for compliance purposes.
Each runner is represented by a card that shows its health status:
A green dot indicates a 'Healthy' state, meaning the runner is operational and executing tasks as expected.
A red dot labeled 'Unhealthy' signals that there are issues that may need investigation or intervention.