Description
Meta is seeking an AI Networking Software Engineer to lead the development of a collective communication library for its large-scale GPU training infrastructure, with a focus on GenAI/LLM scaling. The role involves building customized software features, benchmarks, performance tuners, and software stacks around NCCL and PyTorch to improve distributed ML reliability and performance. Candidates need a bachelor's degree or equivalent practical experience, C/C++ and Python programming skills, a track record of leading projects, and specialized experience in machine learning, deep learning, HPC, GPU architecture, distributed training, or AI infrastructure.
