Senior Software Engineer, NCCL and CUDA - CSP Engagements

• Engage with Cloud Service Providers to identify and address functional and performance issues in NCCL and CUDA libraries
• Analyze and enhance multi-GPU workload performance through profiling and tuning
• Resolve data movement issues in multi-node clusters related to NCCL and NVSHMEM
• Address CUDA porting challenges for customer workloads
• Implement datacenter-specific scheduling for optimal performance
• Debug complex GPU computation and memory-related issues
• Collaborate with customers to tackle workload integration challenges with NCCL and CUDA libraries

• 5-7 years of experience in parallel programming and communication libraries such as MPI, NCCL, and NVSHMEM
• Proficiency with performance optimization tools (e.g., Nsight, nvprof)
• Strong C/C++ programming skills with significant experience in CUDA
• Familiarity with PCIe and NVLINK technologies
• Deep understanding of operating systems and data center architecture
• Knowledge of high-performance networking (InfiniBand, RoCE)
• BS or MS in Computer Engineering, Computer Science, or a related field, or equivalent experience

• Equity participation
• Comprehensive health plans
• 401(k) savings plan
• Generous paid time off and holiday schedule
• Professional development opportunities

Our client is an equal opportunity employer. We encourage you to apply even if you don't meet every qualification-your background could be exactly what this team needs.