@Kenny Shaevel​ :

Yes, there is a hierarchy of permissions within schema permissions in Databricks. Here is the order of permissions, from least to most permissive:

1. USAGE: Allows the user to see the schema and its contents, but not interact with them in any way.
2. CREATE: Allows the user to create objects (e.g. tables) within the schema, but not interact with existing objects.
3. ALTER: Allows the user to modify existing objects within the schema.
4. DROP: Allows the user to delete objects within the schema.
5. ALL PRIVILEGES: Grants all of the above permissions.

Regarding your specific questions:

1. No, the CREATE permission does not implicitly include the ability to read (SELECT) tables within the schema. However, a user with the CREATE permission can create tables with read permissions for specific users or groups.
2. You are correct that the USAGE permission is required for a user to interact with objects within the schema, so the correct answer would be "GRANT USAGE CREATE ON SCHEMA customers TO data.engineer@company.com".
3. Again, you are correct that granting ALL PRIVILEGES would include the SELECT permission, but it would grant more permissions than necessary for the task at hand.

In summary, the best approach would be to grant the team member USAGE and SELECT on the schema, which would allow them to view the tables within the schema without granting unnecessary permissions.