The ML pipeline assumes a curated prefix that a person loads by hand. Official 3.3 knowledge includes automating data ingestion with the same orchestration so training or inference jobs see fresh data without a human copy. What should they add?
Select an answer to reveal the explanation.
Short Explanation
The pipeline assumes a curated prefix a person loads by hand. Wire ingest automation into the pipeline with Glue or an S3 event into SageMaker Pipelines or EventBridge. That is not Domain 1 format design, and a hand copy is the failure.
Full Explanation
Integrate data ingestion into CI/CD or SageMaker Pipelines / EventBridge orchestration (Glue or an S3 event as a pipeline step) so jobs see fresh data without a human copy. That is ingest automation, not Domain 1 format or transform design. Polly is not that step.