Open to opportunities

Michael May

@michaelmay

Message

Staff software engineer architecting low-latency streaming and personalization at scale.

United States

Message

What I'm looking for

I’m looking to lead architecture for real-time streaming, fraud/personalization, and ML feature platforms—delivering low-latency systems with strong data quality, reliable CI/CD, and deep observability in a collaborative environment.

I’m a Staff Software Engineer (L6) on Amazon’s Personalization & Fraud Detection Engineering (PFDE) team, where I led the architecture and development of the Real-Time Event Streaming & Feature Platform for low-latency personalization, fraud detection, and analytics across billions of daily events.

I built and evolved core systems end to end—architecting streaming with Amazon MSK, Kinesis, and Apache Flink; migrating legacy batch ETL to a serverless data lake on Amazon S3 with Apache Iceberg, AWS Glue, and PySpark; and delivering microservices, REST APIs, and real-time dashboards. I also drove ML feature stores and pipelines (Amazon SageMaker, EMR Spark, AWS Lambda), automated data quality checks with Amazon Deequ, and established monitoring/observability (CloudWatch, Grafana, Prometheus, OpenTelemetry). Along the way, I built CI/CD (CodePipeline, CloudFormation, Terraform, Docker, GitHub Actions) and mentored junior engineers to raise operational excellence.

Experience

Work history, roles, and key accomplishments

Current

Staff Software Engineer

Current

Amazon

Apr 2020 - Present (6 years)

Architected a real-time event streaming and feature platform processing 10B+ daily events, reducing data latency from hours to seconds for personalization and fraud detection. Migrated batch ETL to a serverless data lake on S3/Iceberg, improving processing efficiency by 40%, cutting infrastructure costs by 25%, and building CI/CD, monitoring, and automated data quality to reduce incidents by 60%.

Amazon Kinesis Apache Flink Java Spring Boot Pyspark AWS Lambda Terraform Apache Iceberg

Senior Software Engineer

Spotify

Jan 2016 - Mar 2020 (4 years 2 months)

Built A/B testing and personalization infrastructure to enable real-time experiment orchestration and data-driven product decisions. Developed .NET services and dashboards, integrated Kafka event streams for low-latency monitoring, and created pipelines combining C# services with Python ML scripts to improve experiment velocity and data freshness.

.NET C ASP.NET Core Kafka A B Testing TypeScript REST APIs Data Validation Monitoring Python

Software Engineer

Plaid

Apr 2012 - Nov 2015 (3 years 7 months)

Developed Plaid’s Core Banking API platform for secure, real-time account linking and transaction normalization used by fintech clients. Built Ruby on Rails and Python API endpoints, designed normalized transaction schemas, and implemented OAuth, tokenization, encryption, and audit logging; optimized high-throughput performance using Redis and Sidekiq/Resque.