Batch write and discover records in Amazon SageMaker Feature Store

This article announces two new APIs for Amazon SageMaker Feature Store: BatchWriteRecord for efficient bulk ingestion and ListRecords for record discovery and enumeration.

• BatchWriteRecord writes up to 25 records across multiple feature groups in a single API call, reducing throughput bottlenecks from single-record ingestion
• Supports partial-success semantics, per-record TTL control, and EventTime-based ordering guarantees matching PutRecord behavior
• ListRecords enumerates record identifiers within feature groups using pagination, working with both Standard (DynamoDB) and In-Memory (Redis) storage tiers
• Enables record discovery and cleanup for In-Memory tier, previously impossible without offline store fallback
• Supports compliance workflows like verifying complete data deletion across multiple feature groups

These APIs address operational gaps in high-throughput feature pipelines and provide critical record lifecycle management capabilities for all Feature Store storage tiers.