Triton Compiler and Kernel Software Engineer
Core
Develop and optimize the Triton compiler and high-performance GPU kernels for AMD GPUs, enabling AI workloads and multi-GPU communication.
Role type
Senior IC compiler and kernel engineer (GPU/AI)
Builds
Triton compiler, ROCm software stack, high-performance AI kernels
Domain
AI software stack, GPU architecture, compilers
Deliverable
production ML models
Required skills
GPU architecture, compiler development, high-performance kernel optimization, multi-GPU communication, distributed AI training, low-level performance analysis
Preferred skills
AMD GPU/ROCm experience, Triton/HIP/CUDA optimization, LLVM/MLIR, collective communication libraries, distributed training frameworks, AI coding tools
Technologies
Triton, ROCm, HIP, CUDA, LLVM, MLIR, PyTorch, RCCL, NCCL
Responsibilities
Develop and optimize Triton compiler support for AMD GPUs; Improve compiler lowering, optimization, scheduling, and code generation; Create high-performance kernels for attention, GEMM, MoE, and other AI workloads; Develop and optimize multi-GPU kernels and communication primitives; Analyze compute, memory, communication, occupancy, register usage, and generated code; Resolve complex correctness and performance issues across single- and multi-GPU workloads; Provide technical leadership, code reviews, and mentoring
Seniority
Senior, hands-on IC
