Research Oracle before you apply
Check ratings, real-employee reviews, verified pay, and interview difficulty.
This role will lead complex technical programs, establishes scope and milestones, improves cross-team processes, automates reporting, and acts as a technical liaison across stakeholder teams.
Responsibilities
Primary Responsibilities
Program Ownership
Lead one or more major GPU Cluster Health domains such as strategic customer availability, partner repair execution, RMA/spares governance, data/reporting framework, repair workflow improvement, or engineering platform tooling.
Translate availability gaps into structured programs with scope, owners, milestones, KPIs, risks, dependencies, and executive-ready status.
Drive weekly operating rhythm for assigned domains, including KPI review, blockers, escalations, decisions, and follow-through.
Customer and Horizontal Leadership
Support the vertical/horizontal TPM model by owning either a customer vertical or a horizontal functional area.
For vertical ownership, lead availability and repair strategy for assigned strategic customers and customer clusters.
For horizontal ownership, lead cross-cutting programs such as NVIDIA/AMD partner tracking, SDE/SRE tooling, RMA/spares feedback loop, SOP tracking, or metrics/reporting framework.
KPI Definition and Governance
Define KPI measurement methods for repair health, including cluster availability, unavailable-host backlog, repair cycle time, repair success rate, reopen rate, spare availability, RMA loop performance, partner responsiveness, and SLA/SLO adherence.
Drive consistent reporting standards across TPMs and support teams.
Identify trends across multiple workstreams and convert them into prioritized corrective actions.
Cross-Functional Execution
Partner with India, Morocco, and Mexico SDE/SRE teams to identify tooling and first-level support needs.
Prioritize tooling requirements that reduce manual repair coordination, improve triage, automate reporting, or accelerate repair handoffs.
Work with GSL, CHS, CPV, TRS, Warminator, and data center operations to remove repair blockers.
Support incident-style escalation for clusters at availability risk.
Continuous Improvement
Lead post-program retrospectives and drive process updates.
Standardize repeatable playbooks, intake processes, escalation paths, and repair governance artifacts.
Identify automation opportunities and advocate for prioritization.
Expected Outcomes
Major repair programs have clear objectives, metrics, owners, and measurable improvement.
Leadership has reliable visibility into customer cluster health and repair risk.
Repair blockers are resolved through structured governance instead of ad hoc escalation.
Tooling and reporting requirements are translated into actionable SDE/SRE work.