Kevin Zhou

I’m a senior software engineer with 12+ years at Amazon, specializing in large-scale cloud observability and video streaming. I’ve built reliable real-time systems at AWS and Amazon Prime Video that handle billions of events daily while delivering smooth playback for millions of users. My focus is latency, reliability, and operational excellence—end to end.

At AWS, I owned real-time metric ingestion pipelines in Python and Java, integrating Kafka streaming to deliver sub-second latency at massive scale. I designed and optimized REST API and GraphQL endpoints for CloudWatch metrics querying, cutting API response times by 40% under high load. I also helped bring generative AI into CloudWatch Logs Insights to enable natural language query generation from user prompts.

I build for fast diagnosis and continuous improvement: I developed real-time dashboards with React and WebSockets, improved troubleshooting speed, and reduced MTTR by 35%. As L5, I owned observability SLOs/SLIs and implemented anomaly detection with Python and composite alarms, while strengthening secure data flows with IAM auth, encryption, and logging. I lead CI/CD using AWS CodePipeline, CodeBuild, and GitHub Actions—automating testing and deployments and cutting release cycles by 25%—and I mentor teams on unit testing and CI/CD best practices tailored to AWS-scale environments.