Manager, Performance Research and Analysis

Lead end-to-end performance research and execution for next-generation NVIDIA AI GPU clusters powering distributed training and inference. Drive characterization, test plans, and optimizations across RDMA/PRDMA networking, collective communication (NCCL), congestion control, and load balancing. Own performance observability by building telemetry-driven dashboards and automated analytics across NICs, switches, GPUs, and NVLink, and perform deep root-cause analysis to mitigate multi-node bottlenecks.