AI基础设施网络研发专家 - 计算
Core
Designing and developing core container network functions and end-to-end architecture for large-scale Kubernetes, AI training, and GPU clusters.
Role type
Senior IC AI Infrastructure Network Engineer
Builds
High-performance container networks and unified heterogeneous compute scheduling systems for AI-native infrastructure
Domain
Cloud infrastructure, AI/ML, Networking
Deliverable
production ML models
Required skills
Golang, Rust, C, C++, Computer Networking, Kubernetes, CNI, RDMA, RoCE, InfiniBand, eBPF, Network Virtualization, TCP/IP, Routing & Switching
Preferred skills
AI training communication patterns, NCCL, GPU Direct RDMA, Topology-aware scheduling
Technologies
Kubernetes, CNI, eBPF, RDMA, RoCEv2, InfiniBand, NCCL, Calico, Cilium, Flannel
Responsibilities
Design and develop core container network functions and end-to-end architecture; Build AI high-performance container networks including RDMA and heterogeneous compute fusion; Optimize network performance, stability, and scalability for large clusters; Track and drive adoption of technologies in Kubernetes, CNI, eBPF, and AI networking.
Seniority
Senior, hands-on IC
