Staff Software Engineer, SIMD Kernels

What you'll do

• Build and maintain kernels for ML operators such as softmax, layer normalization, and activation functions.
• Map algorithms and AI-framework computational graphs to modern hardware, considering hardware–software co-design trade-offs.
• Develop SDK solutions that help developers use the software stack and analyze performance.

What they ask for

• MS or PhD in computer engineering, math, physics, or a related degree, with 5+ years of industry experience.
• Strong grasp of computer architecture, data structures, system software, and machine learning fundamentals.
• Proficiency in C/C++ and Python development in a Linux environment; experience implementing algorithms in C/C++ and Python.
• Experience implementing algorithms for specialized hardware (such as FPGAs, DSPs, GPUs, or AI accelerators) and implementing operators commonly used in ML workloads.

Nice to have

• Prior startup, small team, or incubation experience.
• Experience with ML frameworks, ML compilers, embedded SIMD vector processors, or work at a cloud provider or AI compute/subsystem company.

Pay

$190K – $300K (base salary)

Tools

C/C++, Python, Linux, CUDA, TensorFlow, PyTorch, MLIR, LLVM

AI focus

Build software kernels for ML operators and AI hardware, supporting AI inference.