Principal Software Engineer, Kernels

What you'll do

• Develop, enhance, and maintain software kernels for AI hardware, mapping algorithms and AI-framework computational graphs to hardware architectures.
• Implement and optimize ML operators and algorithms for specialized hardware, balancing hardware-software co-design tradeoffs.
• Work with compiler, ML, systems, and hardware teams to build compiler infrastructure and scale software deliverables.

What they ask for

• MS in computer engineering, math, physics, or a related degree with 12+ years of industry experience, or a PhD in one of these fields or a related field with 7+ 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 Linux environments; experience implementing algorithms in high-level languages.
• Experience implementing algorithms for specialized hardware, implementing common ML operators, and developing for embedded SIMD vector processors such as Tensilica.

Nice to have

• Prior startup, small team, or incubation experience.
• Experience with ML frameworks or compilers, deep learning models, or work at a cloud provider or AI compute/subsystem company.

Pay

$195K – $285K (base salary)

Tools

C/C++, Python, Linux, FPGAs, DSPs, GPUs, CUDA, Tensilica

AI focus

Build and optimize software kernels and compiler infrastructure for AI inference hardware, including operators used in machine-learning workloads.