IAM / ml / sagemaker

Amazon SageMaker (IAM)

AWSsagemakerml

Amazon SageMaker is a fully managed machine learning service that enables data scientists and developers to build, train, and deploy ML models at scale.

iamaws

docs ↗

Permissions

Actions in this service

sagemaker:ListJobs

sagemaker:DescribeJob

sagemaker:CreateJob

sagemaker:StopJob

sagemaker:ListModels

Policy

Minimal privilege example

Copy-paste least-privilege policy statement — not yet harness-verified.

{ "Version": "2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "sagemaker:ListJobs", "sagemaker:DescribeJob", "sagemaker:CreateJob", "sagemaker:StopJob", "sagemaker:ListModels" ], "Resource": "*" } ] }

SDK mapping

Method → permission

SDK method mapping (e.g. boto3.get_object() → s3:GetObject) is not yet collected.

Footguns

Known gotchas

Avoid sagemaker:* — grants full control including delete and modify of models, endpoints, and jobs

Avoid sagemaker:DeleteModel and sagemaker:DeleteEndpoint unless explicitly needed for cleanup

Resources

docshttps://servicereference.us-east-1.amazonaws.com/v1/sagemaker/sagemaker.json ↗