An ML engineer must monitor a SageMaker model to automatically detect shifts in the input feature distributions. Which approach provides this capability with minimal operational overhead?

The correct answer leverages SageMaker Model Monitor, specifically its data quality monitoring capabilities, to automatically detect shifts in input feature distributions. Establishing a data quality baseline allows Model Monitor to compare incoming inference data against expected distributions. Enabling emitmetrics ensures that Model Monitor publishes relevant data quality metrics to CloudWatch, which can then be used to trigger alarms for automated notifications. This approach offers minimal operational overhead as Model Monitor handles the data collection, analysis, and metric generation. Incorrect options: "Model quality baseline" is used for monitoring model performance (e.g., accuracy, precision), not input feature distributions. SageMaker Debugger is primarily for debugging training jobs, not for monitoring deployed model inference data for distribution shifts. Manually analyzing CloudWatch logs for data drift is not automated and would incur significant operational overhead, failing the "minimal operational overhead" requirement.