AI网络研发工程师（J79245）
Core
Diagnose and optimize network performance, develop and deploy communication libraries, and accelerate training/inference for Baidu's AI computing clusters.
Role type
Senior IC network software engineer (AI infrastructure)
Builds
High-performance communication libraries and network optimization tools for AI training and inference
Domain
AI computing infrastructure / High-performance networking
Deliverable
production ML models
Required skills
C/C++, TCP/IP protocol design, Linux kernel internals, GPU programming, RDMA, collective communication libraries
Preferred skills
Shell scripting, Python, system performance profiling, bottleneck optimization
Technologies
Linux, GPU, RDMA, C++, Python, Shell
Responsibilities
Diagnose and optimize network performance in AI computing clusters; Develop and maintain communication libraries; Implement communication acceleration for training and inference; Develop associated software for cluster networking
