Meta
Software Engineer, SystemML - AI Networking
$154K–$217K
Menlo Park, CA · Full-time · Posted Aug 21, 2026
- Experience
- No years specified.
- Education
- Bachelor’s degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience.
About the job
Leads development of Meta’s collective communication software stack to improve the reliability and performance of large-scale distributed GPU training, especially GenAI and LLM workloads.
What you'll do
- Provide technical leadership for collective communication library development on large-scale GPU training infrastructure, focusing on GenAI/LLM scaling.
- Develop and improve software features, benchmarks, performance tuners, and stacks around NCCL and PyTorch.
What they ask for
- Bachelor’s degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience.
- Proven C/C++ and Python programming skills and a track record of leading successful projects.
- Experience leading cross-functional technical projects and communicating technical decisions to technical and non-technical stakeholders.
- Specialized experience in at least one listed ML/deep-learning domain; experience with NCCL and distributed GPU performance analysis on RoCE/Infiniband.
Nice to have
- Knowledge of GPU architectures and CUDA programming; experience with PyTorch, Caffe2, or TensorFlow.
- PhD in Computer Science, Computer Engineering, or a relevant technical field; experience in HPC and parallel computing.
Pay
$154,003/year to $217,000/year (base salary)
Tools
C/C++, Python, NCCL, PyTorch, RoCE, Infiniband, CUDA, Caffe2, TensorFlow
AI focus
Builds and leads software for reliable, high-performance distributed GPU training and communication, with a focus on GenAI/LLM scaling. The posting also asks for ongoing AI skill development, use of AI tools to improve workflows, and responsible AI practices.
Model-written summary · the original posting is authoritative