Principal Developer, AI Networking
Core
Profile, analyze, and optimize AI workloads on large-scale GPU and CPU clusters for distributed Deep Learning LLM training and inference, with a primary focus on collectives communication and networking.
Role type
Principal Developer, AI Networking
Builds
High-performance network systems and performance analysis tools for LLM workloads
Domain
AI Networking / High-Performance Computing
Deliverable
production ML models
Required skills
High-performance networking (RDMA, MPI, NCCL, SHARP), Performance evaluation techniques, NVIDIA GPUs and CUDA, Deep learning frameworks (TensorFlow, PyTorch), Networking collective communication libraries (NCCL, RoCE), Python, Bash, C++, Container-based development
Preferred skills
AI workload benchmarking for distributed LLM training, PyTorch/CUDA/NCCL expertise, Comprehensive system knowledge (Intel/AMD/ARM CPUs, NVIDIA GPUs, HCA, Memory, PCI), Contemporary performance evaluation tools
Technologies
PyTorch, CUDA, NCCL, RoCE, RDMA, MPI, SHARP, Python, Bash, C++
