L3 Technical SME - AI Compute
Core
Senior technical authority for AI infrastructure platforms, providing deep expertise across GPU compute, Linux systems, AI cluster operations, and high-performance infrastructure environments.
Role type
Senior IC AI Infrastructure Engineer (GPU Compute)
Builds
Mission-critical AI environments, large-scale GPU platforms, and high-performance computing clusters.
Domain
AI Infrastructure / High-Performance Computing / Data Center Hardware
Deliverable
production ML models
Required skills
Linux systems administration, NVIDIA GPU diagnostics, PCIe/NVLink/NVSwitch architecture, Kubernetes/Slurm orchestration, root cause analysis, firmware lifecycle management, capacity planning, incident management, OEM/vendor engagement, automation scripting.
Preferred skills
Python/Ansible/Bash automation, SRE principles, NVIDIA DGX/HGX platforms, liquid-cooled infrastructure, DCGM/NVIDIA-SMI telemetry, InfiniBand/RDMA networking, Mandarin language proficiency.
Technologies
Linux, NVIDIA GPUs, Kubernetes, Slurm, Python, Ansible, Bash, PCIe, NVLink, NVSwitch, BMC, Redfish, IPMI, InfiniBand, RDMA, DCGM, NVIDIA-SMI.
Responsibilities
Act as highest level technical escalation for complex AI compute platform incidents; Lead advanced fault isolation and troubleshooting across GPU/CPU/memory/firmware layers; Perform deep analysis of GPU health, XID events, ECC errors, and performance degradation; Lead technical root cause analysis (RCA) and define permanent corrective actions; Drive platform lifecycle management including firmware validation and change risk evaluations; Automate operational tasks and remediation workflows; Analyze capacity, utilization, and SLOs to recommend optimization opportunities.
Seniority
Senior, hands-on IC
