Senior Software Architect - Deep Learning and HPC Communications

Tasks

• Collaborate on hardware and software co-design
• Conduct experiments
• Design communication technologies
• Develop proofs of concept
• Identify communication bottlenecks
• Implement communication technologies
• Perform quantitative modeling
• Use simulation to evaluate GPU cluster performance

Perks/Benefits

• Equity
• Health benefits

Skills/Tech-stack

C# | C++ | CUDA | CUDA programming | Communication Scaling | Computer Architecture | Deep learning | Infiniband | Linux | MPI | NCCL | NVLink | NVSHMEM | Network Architecture | Network Topology | OpenSHMEM | Operating System | Operating System Concepts | Parallel Programming | PyTorch | RoCE | System Concepts | TensorFlow | UCC | UCX