Advanced Micro Devices-posted 3 months ago
Hybrid • San Jose, CA
5,001-10,000 employees
Computer and Electronic Product Manufacturing

At AMD, we care deeply about transforming lives with our technology to enrich our industry, our communities, and the world. Our mission is to build great products that accelerate next-generation computing experiences - the building blocks for the data center, artificial intelligence, PCs, gaming, and embedded systems. Underpinning our mission is the AMD culture, where we push the limits of innovation to solve the world's most important challenges. We strive for execution excellence while being direct, humble, collaborative, and inclusive of diverse perspectives. The Triton Compiler Engineer role involves working on Triton, a language and compiler for writing highly efficient custom machine learning GPU kernels, which is widely adopted in open AI software stack projects like PyTorch, vLLM, SGLang, and many others. AMD GPU is an official backend in Triton and of strategic importance to AMD. We are fully committed to making it awesome. If you are interested in making GPUs run fast by developing the Triton compiler and kernels, please join us!

  • Develop and maintain Triton compiler's AMD backend
  • Improve various compilation patterns and passes in Triton
  • Research and author high performance matmul and attention kernels in Triton
  • Profile kernel performance on AMD GPUs and improve bottlenecks
  • Fix issues in AMDGPU backend in LLVM
  • Familiarity with compilers, GPU architectures, parallel programming, and/or high-performance kernels
  • Comfortable performing quantitative analysis of workload and driving improvements across different compiler layers
  • Willingness to learn and work across boundaries
  • Familiarity or existing experience with Triton
  • Familiarity with compiler projects like MLIR, LLVM, IREE
  • Deep understanding of GPU architectures and programming models
  • Deep experience with writing high performance GPU kernels and GPU performance tuning
  • Experience debugging cross-stack issues and reducing user problems to actionable execution
  • Open-source development ethos
  • AMD benefits at a glance
© 2024 Teal Labs, Inc
Privacy PolicyTerms of Service