run-experiment

You can run machine learning experiments on a remote GPU server by using this Skill to synchronize code via rsync or git, validate GPU availability, and launch the job in an isolated screen session for secure remote execution.

The best way to sync code and launch training is using rsync or git for transfer, followed by launching the experiment in an isolated screen session. This ensures efficient deployment and resource isolation on remote GPU servers.

Yes, you can execute local GPU training jobs in the background without blocking the terminal. This Skill launches local experiments as background processes, ensuring resource isolation and maintaining terminal responsiveness while training runs.

Yes, this remote execution workflow supports both git and rsync for code synchronization. It detects the target environment by reading CLAUDE.md and uses these tools to efficiently transfer your machine learning code to remote GPU servers.

You validate GPU availability before deploying a model training job by letting this Skill check the specified compute environment. It automatically verifies GPU presence on local or remote servers before activating the environment and launching the experiment.