GPU Software Engineer/GPU Architect

Dice is the leading career destination for tech experts at every stage of their careers. Our client, Triune Infomatics Inc, is seeking the following. Apply via Dice today!Role: GPU Software Engineer/GPU ArchitectLocation: San Jose, CADuration: Long-term >> ongoing contractOverview: We're looking for a strong GPU Software Engineer/GPU Architect to join a highimpact engineering team working on nextgeneration AI, GPU, and semiconductor technologies. This role focuses on GPU kernel development, memory architecture, and integration with modern inference systems such as vLLM and SGLang. You'll work onsite in San Jose, collaborating closely with a team of engineers building highperformance GPUaccelerated systems.Develop and optimize CUDA/ROCm kernels for AI workloadsWork with HBM, memory hierarchy, thread scheduling, and P2P communicationIntegrate GPU kernels with vLLM, SGLang, and other inference serversBuild highperformance components in C++ and PythonSupport AI frameworks such as PyTorch and TensorFlowOptimize multiGPU scaling, KVcache, and attention kernelsProfile and debug GPU workloads using Nsight, rocprof, etc.Collaborate with crossfunctional GPU, AI, and semiconductor teamsRequired Skills:Strong experience with CUDA, ROCm/HIP, OpenCL, or MPIDeep understanding of GPU architecture, HBM, memory models, and thread hierarchiesHandson experience with AMD/NVIDIA GPU software stacksExpertlevel C++ and PythonExperience with PyTorch or TensorFlowExperience with vLLM, SGLang, or similar inference systemsPreferred Skills:RDMA, RoCE, InfiniBand, or Infinity FabricDistributed inference/training or HPC experienceSemiconductor or hardwareadjacent experience