Question

A retail company runs a real-time SageMaker endpoint for product recommendations. Business analysts report that recommendation click-through rates have gradually declined over the past three months, though the endpoint shows no errors and stable latency. The ML team confirms input feature distributions have shifted as customer behavior evolved seasonally. The team wants an automated, low-maintenance approach to keep the deployed model accurate over time without manually inspecting metrics each week. Which approach best addresses this requirement?

Accepted Answer

Enable SageMaker Model Monitor model quality monitoring with a ground-truth ingestion pipeline, and configure a CloudWatch alarm on the monitored accuracy metric to trigger an EventBridge rule that starts a retraining pipeline.

Answer

Increase the endpoint instance count and enable target-tracking auto scaling so the model can handle more traffic as customer behavior changes.

Answer

Enable SageMaker Model Monitor data quality monitoring only, and rely on the endpoint's built-in latency and invocation CloudWatch metrics to detect the decline.

Answer

Switch the endpoint to a serverless inference configuration so the model automatically adapts to the changing feature distributions.