Amazon SageMaker is a fully-managed AWS service that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale.
Amazon SageMaker is Amazon's official system for developing machine learning systems in the cloud. It is based on Jupyter notebooks and has the ability to talk with other AWS services in order to explore data, automate processes and conduct various kinds of analysis.
It provides ready-to-use frameworks and algorithms to deal with different use cases and gives the ability to develop fully custom models.
One of the most important capabilities of SageMaker is the ability to deploy models in fully scalable and secure environments.
The only prerequisites for using Amazon SageMaker are an AWS account and an IAM (Identity and Access Management) admin user.
Tagging Recommendation:
Use the 
amazon-sagemaker tag for all SageMaker-related questions.
If it is a question about amazon-sagemaker-studio must also be redirected to this tag as a subset.
For all other SageMaker sub-services, if there is not yet a dedicated tag, it is good to redirect them to SageMaker only if they are closely related.
FAQ:
There are some recurring and important questions that one is faced with when starting to use SageMaker that should not be duplicated:
- Load S3 Data into AWS SageMaker Notebook
- How to use SageMaker Estimator for model training and saving
- How to schedule tasks on SageMaker
- AWS Sagemaker - Install External Library and Make it Persist
- How to pass dependency files to sagemaker SKLearnProcessor and use it in Pipeline?
References
- What Is Amazon SageMaker?
- Official documentation for SageMaker - Boto3: 1.25.4
- Amazon SageMaker Python SDK
