About the role
Job Details: Job Description: Join us to help build the next generation of AI hardware solutions. You will be part of a highly skilled, agile team developing cutting-edge hardware for the AI domain, where we push the boundaries of what silicon can do for emerging AI workloads. With a startup-like culture, we move quickly and give engineers the opportunity to drive significant technical and business impact. We are continuously developing modern and effective working methods, including hands-on adoption of AI tools throughout the chip development flow. Mission: Define and own the pod-level reliability specifications that ensure the availability, resilience, and serviceability of a large-scale data center across hardware, thermal, and operational dimensions.
Responsibilities: Define and maintain pod-level reliability/availability specs and targets (MTBF, AFR, RAS) for compute, memory, storage, network, power, and cooling subsystems. Translate system/SLA
requirements into pod and subsystem level reliability specs; flow
requirements down to silicon, platform, and facilities teams. Lead FMEA, root-cause analysis, and pod fleet failure-data analytics to drive corrective actions and spec updates. Architect RAS features (ECC, memory mirroring, predictive failure, telemetry) and graceful degradation/redundancy against pod-level specs. Partner with facilities on pod power/cooling redundancy (N+1, 2N), thermal margins, and disaster-recovery readiness. Establish HALT/HASS, burn-in, qualification processes; track field returns and KPIs against pod spec.
Qualifications: Minimum
Qualifications: BS/MS/PhD in EE/ME Reliability or related; and/or at least 4-6 yrs experience. Experience authoring and owning reliability specs and requirement flow-down. Strong RAS, FMEA, statistical reliability (Weibull, FIT) skills. Experience with large-scale fleet telemetry and thermal/power redundancy. Preferred