SageMaker AI (Canvas) - Shared Space

Yes, creating a shared space would be ideal for your situation where multiple users need to share their work with each other in SageMaker.

A shared space in Amazon SageMaker AI consists of a shared JupyterServer application and a shared directory. This enables real-time collaboration between users in your domain, allowing them to access, read, edit, and share notebooks in near real-time. All user profiles in a domain have access to all shared spaces within that domain.

To create a shared space:

1. First, ensure your domain supports shared spaces. If you created your domain without shared space support, you'll need to add this capability to your existing domain.

2. Designate a space default execution role that will govern the permissions for any user utilizing the space.

3. You can create the shared space through:

• The SageMaker AI console
• Amazon SageMaker Studio
• AWS CLI

Shared spaces are available in both Amazon SageMaker Studio Classic and JupyterLab. They support collaboration features like automatic tagging, real-time co-editing of notebooks, and customization.

Once created, all users in your domain will be able to access the shared space, where they can collaborate on notebooks, files, experiments, and models. This provides the quickest path for your team to start iterating together on their machine learning projects.
Sources
Multiple domains and shared spaces - SageMaker Studio Administration Best Practices
Collaboration with shared spaces - Amazon SageMaker AI
Create a shared space - Amazon SageMaker AI
Amazon SageMaker AI Features - Amazon SageMaker AI