Ok thanks, I’m searching a way to not be dependent on the aws auto scaler script but rather use a k8s job which will run and die after it’s finished thus saving resources

Some way more k8s native like a job with a base image and the exported task

I don’t wan’t to be dependent on the agents

Well, you can create an image using clearml-agent build , but you need the agent since that's the piece of code responsible for everything related to running tasks (i.e. building the environment, fetching the code, patching any code diffs, obtaining task metadata from the server, reporting on task progress and status, etc.)

Or my own computer

Hi ScrawnyLion96 , how would you like to trigger that, conceptually?