Amazon SageMaker Feature Store is the correct choice because it provides a centralized and governed repository for machine learning features with built in versioning and both online and offline stores to ensure consistency between training and inference.

The Feature Store lets teams register discover and reuse features across squads and supports point in time access access control and encryption to meet compliance and production needs. It offers an online store for low latency inference and an offline store for batch training so features remain consistent across development and production workflows.

Amazon SageMaker Data Wrangler focuses on data preparation and transformation and it does not act as a feature registry or provide built in feature versioning for cross team reuse.

Amazon SageMaker Ground Truth is a labeling and annotation service and it is unrelated to storing or sharing engineered features for model training and inference.

Amazon SageMaker Canvas is a no code modeling environment and it does not manage a centralized versioned feature store for multiple data science teams.

Full AWS Practitioner Certification Question