- New York, New York, United States
- Staff
- $200K–$280K per year
- 5–8 years
Compensation
$200,000—$280,000 USD
Responsibilities
- Own the strategy and roadmap for a significant Fleet Operations product area
- Lead multi-sprint, cross-functional initiatives from problem framing through rollout across live GPU clusters
- Turn operational ambiguity into product: shadow on-call rotations, ride along with support and repair workflows, and translate recurring toil into tooling, automation, and platform capabilities
- Define the metrics that matter for a GPU fleet — availability, utilisation, MTTR, time-to-bring-up, hardware failure rates, support ticket deflection — and drive the roadmap against them
- Partner with engineering on architecture and trade-offs for systems that span bare metal, orchestration, observability, and control planes
Requirements
- Nice to have: degree in computer science
- Nice to have: engineering
- Nice to have: SRE
- Nice to have: bare-metal provisioning
- Nice to have: OpenStack Ironic