异构云服务器高级技术专家-计算
Core
Design and develop heterogeneous cloud server architecture, optimize performance, ensure stability, and lead unified scheduling for compute resources.
Role type
Senior IC heterogeneous cloud server technical expert
Builds
Heterogeneous cloud servers (GPU/NPU/CPU/RDMA) and unified compute resource scheduling systems
Domain
Cloud Infrastructure / High-Performance Computing / AI Hardware
Deliverable
production ML models
Required skills
C++/Python/Go, Linux Kernel, GPU/NPU/CPU hardware architecture, RDMA networking, system debugging, distributed computing, collective communication, CUDA, AI performance tuning, Kubernetes, containerization, virtualization (QEMU/KVM)
Preferred skills
Development of performance analysis and fault troubleshooting tools for GPU/NPU/CPU, experience with AI performance optimization
Technologies
Linux Kernel, CUDA, Kubernetes, QEMU, KVM, RDMA
Responsibilities
Design and develop heterogeneous cloud server architecture, optimize server performance, design stability architecture and monitoring systems, participate in unified scheduling scheme design for heterogeneous compute resource pools
Seniority
Senior, hands-on IC
