Hardware Health
Core
Design and develop hardware health monitoring, predictive analytics, and autonomous remediation systems for exascale-class GPU clusters to ensure reliability and performance.
Role type
Senior IC hardware reliability engineer (GPU clusters)
Builds
Predictive health models, failure detection frameworks, and real-time observability platforms for AI training infrastructure
Domain
AI infrastructure / High-performance computing / GPU systems
Deliverable
production ML models
Required skills
GPU architecture, high-speed interconnects (NVLink, InfiniBand, RoCE), hardware telemetry, failure analysis, reliability modeling, Python, C++, C#, Java, JavaScript
Preferred skills
Exascale systems experience, machine learning-based anomaly detection, predictive maintenance, large-scale infrastructure operations
Technologies
NVIDIA H100, GB200, NVLink, NVSwitch, InfiniBand, RoCE
Responsibilities
Design diagnostic frameworks for large GPU clusters; Build predictive analytics pipelines using telemetry/power/thermal data; Collaborate with silicon/firmware teams on root cause analysis; Define system health KPIs and integrate into observability platforms; Lead incident triage for hardware/network/cooling issues; Drive automation to reduce manual intervention; Partner on hardware design for reliability and thermal efficiency
Seniority
Senior, hands-on IC