[12/2/2021 Update: This article contains information about Gradient Experiments. Experiments are now deprecated, and Gradient Workflows has replaced its functionality. Please see the Workflows docs for more information.]
We're delighted to announce the arrival of new secrets management in Gradient!
Secrets are a way to scope keys and credentials locally at the team, cluster, or project level – and they're now available for all Gradient experiments run on a private cluster!
Secrets are often used to store API keys, S3 credentials, SSH keys, or any other secure text needed to run your machine learning workloads. They are especially for collaborative teams in need of secure methods of credentialing outside resources for use in a project or experiment.
Once you've created a secret, simply inject the secret as an environment variable using the syntax
Secrets can be created in the Gradient console or via the Gradient CLI.
For more information including CLI usage, read the docs!