Distributed LLM performance engineering powers Huawei Canada’s Ascend NPU training systems with MindSpeed, hybrid parallelism, and profiling expertise.This Senior System Engineer position focuses on foundation-model pre-training and post-training at scale. You will architect distributed pipelines using MindSpeed, Megatron, PyTorch FSDP, veRL, vLLM, and Ray. The work targets MFU, compute-communication overlap, rollout pipelining, and reliable numerical behavior.Key Responsibilities:
Scale pre-training and RL post-training pipelines across NPU clustersOptimize hybrid TP, PP, CP, EP, and ZeRO parallelismTune RL rollout and actor pipelining for training efficiencyImplement FP8, MXFP8, and mixed-precision training schemesEliminate pipeline bubbles, stalls, and HCCL communication overheadRequirements:
Three-plus years in HPC, accelerator operators, or distributed trainingPractical distributed training experience with listed frameworksProficiency in C/C++ and up-to-date Python systems programmingHands-on PyTorch custom C++ or CANN extension developmentPerformance profiling experience on AI hardwareBuild scalable Ascend NPU training systems by combining distributed frameworks, parallelism tuning, profiling, and precision alignment.