Software Engineer, Hardware Health
OpenAI, San Francisco
The role
On the Hardware Health and Observability team, you’ll build critical infrastructure that keeps OpenAI’s largest compute clusters healthy and operational at scale.
Even small numbers of unhealthy systems can impact large-scale training and inference workloads. This team focuses on minimizing downtime, improving fleet efficiency, and ensuring compute resources remain continuously available to researchers and product teams.
Engineers on this team own problems end-to-end, from defining health signals and debugging failures to building automated remediation systems that operate across millions of GPUs globally.