AI通信库研发工程师 - 网络
Core
Design, develop, and optimize collective communication libraries (CCL) for high-performance multi-card/multi-machine AI clusters, focusing on RDMA-based interconnects and communication-computation co-optimization.
Role type
Senior IC distributed systems engineer (AI communication libraries)
Builds
High-performance collective communication libraries (CCL) and tools for AI training/inference clusters
Domain
AI infrastructure, high-performance computing, distributed systems
Deliverable
production ML models
Required skills
Distributed systems, AI cluster architecture, RDMA technology, collective communication algorithms (Ring, Tree, Halving-Doubling), CCL implementation principles, performance tuning, fault diagnosis, Triton
Preferred skills
Communication-computation fusion, complex scenario adaptation, architecture evolution
Technologies
RDMA, Triton, NCCL
Responsibilities
Design and optimize CCL for multi-node scenarios; Develop high-speed communication libraries based on RDMA; Analyze and optimize collective communication performance (AllReduce, All2All); Solve communication issues in distributed training/inference; Collaborate with training/inference teams to optimize AI model communication efficiency.
Seniority
Senior, hands-on IC
