Scott Laird
What do you do when you're new to a service and all you know is that you're spending huge amounts of money on it and no one is happy with the service's performance? You use science, of course!
The speaker joined a team with a severe OpenSearch performance problem and applied basic monitoring principles, built models to understand the problem space, conducted experiments to understand what was happening under the hood of a managed service, and then halved the system's latency, cut costs by more than half, and left the team with a framework for further improvement.
Scott worked as an SRE at Google for 17 years, working on many products including Chrome, Google Docs, Calendar, and storage in Google Cloud, but never search. More recently he worked as a part of Figma's Production Engineering team.
He lives in the Seattle area and holds strong opinions on monitoring, sources of truth, and Terraform.
author = {Scott Laird},
title = {The Search for Speed},
year = {2025},
address = {Santa Clara, CA},
publisher = {USENIX Association},
month = mar
}