Columbus AWS (Amazon Web Services) Meetup

Customer identity resolution becomes increasingly complex as organizations scale across multiple systems, regions, and data formats. Traditional rule-based approaches often fail to keep up with data variability, require constant manual tuning, and struggle with real-time processing needs.

This session presents a practical approach to building a scalable identity resolution pipeline using AWS services and modern AI techniques. The architecture combines data ingestion through Amazon S3 and AWS Glue, transformation pipelines using Spark on EMR, and machine learning models deployed via SageMaker for entity matching and standardization. Graph-based relationship modeling is implemented using Amazon Neptune to improve resolution accuracy by incorporating household and shared attribute context.

We will walk through how machine learning models can be used for name and address normalization, how intelligent blocking strategies improve matching efficiency, and how feedback loops can be introduced to continuously improve accuracy. The session also highlights how serverless components such as AWS Lambda can be used for orchestration and real-time processing.

SPEAKER BIO
Mosaic Syed is a Senior Data Engineering and Cloud Solutions Architect with over 20 years of experience designing and delivering scalable, secure, and high-performance data solutions across global enterprise environments.
https://www.linkedin.com/in/mosaic-basha-syed-92300856

Midwest Community Day Midwest is coming this June 24th in Indianapolis, Indiana!!!

The Midwest Community Day conference is a one day event where cloud enthusiasts come together to learn, network, share, and celebrate our shared interests. Sessions cover all aspects of AWS organized into 3 or more concurrent tracks.