Positron, Posit’s integrated development environment for data science, can now run inside Amazon SageMaker AI. The setup combines R and Python development with AWS data, managed training and deployment resources, reducing the need to move a project among separate local and cloud tools.
Administrators build Posit’s container definition, store it in their own Elastic Container Registry and attach it to a SageMaker Studio domain. Data scientists then launch Positron in a Studio Space under that Space’s execution role. That identity controls access to Athena, the Glue Data Catalog and S3, so users do not need to store separate cloud keys inside the editor. Shared Spaces support collaboration, while separate Spaces can isolate projects or compute sizes.
Posit Assistant can use Amazon Bedrock as its model provider, keeping requests within the customer’s AWS account and selected region. The documented example used a synthetic 50,000-loan dataset, moved between R exploration and Python model training, and deployed an XGBoost model to a SageMaker endpoint. Its performance figures describe that demonstration, not a production lending benchmark.
The integration is not a one-click feature for every SageMaker user. Organizations need a Posit license, access to the image definition and administrators who can configure roles, registries and Studio images. For equipped teams, it creates one governed workspace from data inspection through deployment and reporting.