高性能网络通信库研发工程师
Core
Design and implement high-performance collective communication operators (AllReduce, AllGather, etc.) and compute-communication overlap strategies for large model training and inference.
Role type
Senior IC high-performance computing engineer (AI communication library)
Builds
High-performance communication libraries and optimized kernels for large-scale AI training/inference
Domain
AI infrastructure / High-performance computing / Distributed systems
Deliverable
production ML models
Required skills
C++, CUDA, Python, GPU architecture, kernel optimization, distributed training strategies, MPI, NCCL, RDMA verbs
Preferred skills
Large model architecture knowledge (Transformer, GPT, LLaMA), compute-communication overlap design
Responsibilities
Design and implement efficient collective communication operators; Design compute-communication overlap strategies; Optimize operator performance across hardware platforms; Participate in overall library architecture design
Seniority
Senior, hands-on IC
