Jumio builds real-time feature store on AWS with SageMaker

What happened

AWS Machine Learning published a case study on how Jumio built a centralized, real-time feature store on AWS. The architecture uses Amazon SageMaker Feature Store, Amazon Managed Service for Apache Flink, and Amazon Kinesis Data Streams. It delivers sub-100ms feature serving for fraud detection. Jumio saves approximately $120,000 annually with this solution.

Summary written by AI strictly from the sources below

Timeline

1. 17:05
   AWS Machine Learning reported on Jumio's real-time feature store architectureAWS Machine Learning

Sources