Coiled helps individuals and teams manage their resources, control costs, and collaborate with one another. Team members within an account can create software environments, etc. within that account to easily share resources and consolidate billing.

Accounts & Teams#

Upon signing up for Coiled, an account is automatically created for your user, and the name of the account is the same as your username. For example, if you sign up with the username awesome-dev, then we create an account for you named awesome-dev.

If you want to work with a team of two or more users as part of Coiled Pay As You Go or Coiled Enterprise, you can either:

  1. Add other Coiled users to your existing account by using the Team page at<YOUR-ACCOUNT-NAME>/team

  2. Reach out to us at to create an additional account to use for your team such as<YOUR-COMPANY-NAME>/team

Taking the screenshot below as an example, note that this user Kris (seen in the avatar on the top right) is viewing the Team page of the Coiled account (seen in the dropdown on the left).


Working with other accounts#

If you are a team member of another account, you can create clusters or any other resources in the other account by using the account= keyword argument that most of our API commands accept, or by specifying my-team-account-name/ as a prefix in the name= keyword argument.

For example, to create a cluster in the other account:

import coiled

cluster = coiled.Cluster(n_workers=10, account="my-team-account-name")

Or, to create a software environment in the other account:

import coiled

    pip=["dask[complete]", "xarray==0.15.1", "numba"],

You can also configure a different account to act as your default account via your local coiled configuration file.


Software environments which belong to an account are visible and accessible to all account members. This allows team members to easily control, share, and collaborate on their teams’s resources.

Resource limits & tracking costs#

Administrators for each Coiled account can set resource limits for account members like the number of cores a user can allocate at one time or whether or not to grant access to GPUs (which can be expensive). Additionally, you can track each cluster’s cost over time.