Manager, Performance Research and Analysis
Core
Drive end-to-end performance strategy, characterization, test plans, and optimization for next-generation NVIDIA AI GPU clusters, focusing on large-scale distributed training and inference workloads.
Role type
Senior engineering manager (performance research and analysis)
Builds
Next-generation NVIDIA data center solutions and supercomputers based on GPU systems, NIC, Switch, DPU and Networking technologies
Domain
High Performance Computing (HPC) / AI Infrastructure / Networking
Deliverable
production ML models
Required skills
High Performance Networking, RDMA, Systems level performance, Collective communication (NCCL, MPI), Root-cause analysis, Cross-team leadership, Telemetry pipeline design
Preferred skills
NCCL/RDMA/RoCEv2 optimization for multi-thousand GPU deployments, Adaptive routing, PFC/ECN congestion control, AI-assisted root cause analysis agents, Custom Grafana plugin development
Technologies
NCCL, MPI, RDMA, RoCEv2, Grafana, PromQL, LogQL, NVLink, DPU, NIC, Switch, GPU
