Skip to content
Back to jobs

Posted 4 months, 2 weeks ago

AMD

GPU Performance Software Development Engineer

Apply View on Hacker News Company site

Roles

GPU Performance Software Development Engineer

Tech stack

CUDA GPU intrinsics MMA rocprof Attention HIP GEMM GPU compiler pipelines MFMA profiling tools Mixture of Experts (MoE) NVIDIA Nsight Python-embedded DSL MLIR dialects inline assembly MLIR

Location

San Jose, CA

Work setup

Employment
Full-time Employee
Level
Senior

Description

Building Wave, a high-performance GPU programming language and compiler for machine learning workloads. Own end-to-end performance of Wave's kernels. Work includes hand-tuned GPU kernels (GEMM, Attention, MoE, decoding); instruction-level optimization (memory, registers, scheduling, wave/warp behavior); HIP/CUDA, GPU intrinsics (MFMA/MMA), and inline assembly; MLIR dialects, lowering pipelines, and performance-critical compiler passes; profiling via disassembly, counters, rocprof/Nsight. Wave codebase: https://github.com/iree-org/wave. Job posting: https://careers.amd.com/careers-home/jobs/77039?lang=en-us. Location: San Jose, CA.

Similar jobs

  • Loading similar jobs...