Senior Solutions Architect, AI Performance Engineering
Core
Accelerate Physical AI workloads for Autonomous Vehicles and Robotics customers by developing and optimizing high-performance GPU algorithms, data structures, and distributed communication strategies.
Role type
Senior Solutions Architect (AI Performance Engineering)
Builds
Next-generation NVIDIA hardware/software architectures, optimized GPU libraries, and efficient distributed transformer workloads.
Domain
Autonomous Vehicles, Robotics, High-Performance Computing (HPC), Physical AI
Deliverable
production ML models
Required skills
GPU compute efficiency optimization, parallel algorithms, C/C++/Python, Linux, linear algebra, numerical methods, parallel programming, HPC, GPU programming, distributed communication optimization
Preferred skills
CUDA kernel development, Nsight System/Compute, full-stack system evaluation (LLM/HPC), operator/framework/algorithm-level optimization
Technologies
NVIDIA GPUs, cuDNN, cuBLAS, CUTLASS, DeepGEMM, FlashMLA, FlashAttention, Flashinfer, NCCL, NCCL GIN, NVSHMEM, DeepEP, NCCL EP, NVLINK, InfiniBand, RoCE
Responsibilities
Engage with AV application engineers to understand problems and develop parallel algorithms; Collaborate with architecture and research teams to influence next-gen software platforms; Optimize high-performance operators (GPU kernels, instruction tuning, compiler optimization); Improve communication for distributed transformer workloads (interconnects, network protocols, compute-communication overlap).
Seniority
Senior, hands-on IC
