jobapplier

MAX Kernel Engineering Manager

Modular

Skills

Required

  • kernel
  • Triton
  • CUTLASS / CuTe
  • GPU/accelerator kernels
  • CUDA
  • HIP / ROCm
  • NVIDIA
  • AMD / NPUs / ASICs / edge / on-device
  • Mojo
  • tile-level programming models
  • TileLang

Preferred

  • agentic kernel development
  • GPU
  • profiling
  • benchmarking
  • roofline analysis
  • cross-functional communication
  • HPC / ML

Join our team as the Engineering Manager of the MAX Kernel team. In this role, you will lead a team of kernel engineers, own the kernel roadmap, and optimize performance for various hardware platforms. You will collaborate closely with other teams and help grow a kernel engineering community. Additionally, you will be responsible for hiring, coaching, and developing your team, as well as advancing AI-assisted kernel development.

Responsibilities

  • Lead and manage a team of kernel engineers, overseeing their career development, performance management, and team health.
  • Own the kernel roadmap and delivery, ensuring performance optimization for various models and hardware targets.
  • Collaborate cross-functionally with other teams to advance AI-assisted kernel development and the kernel engineering community.

Deep understanding of profiling, benchmarking, and roofline analysis

  • Strong cross-functional communication; able to drive technical decisions with tech leads and communicate status, risks and tradeoffs to engineering leadership and customers
  • 3+ years as an engineering manager leading GPU kernel, compiler, HPC, or ML performance engineering teams. (minimum requirement)

Requirements

  • Track record of hiring, retaining and growing senior kernel or performance engineers
  • Working knowledge of writing and optimizing GPU/accelerator kernels (CUDA, HIP/ROCm, Triton, CUTLASS/CuTe or similar)

Open-source contributions to kernel libraries or inference engines

  • Multi-target experience beyond NVIDIA: AMD, NPUs/ASICs or edge/on-device
  • Experience with Mojo or other kernel and tile-level programming models (Triton, TileLang, CuTe)

Preferred

  • Experience with AI-assisted or agentic kernel development

Pay

USD 255,400 - 383,200