Description

You will develop, optimize, and maintain kernel-level software for AI and machine learning workloads. You design core components for AI compute systems and ensure high-performance execution on specialized hardware.

Responsibilities

• Implement algorithms for specialized hardware including FPGAs, DSPs, GPUs, and AI accelerators using CUDA.
• Design and implement ML operators such as GEMMs, convolutions, BLAS, softmax, layer normalization, and pooling.
• Develop software for embedded SIMD vector processors like Tensilica.
• Debug, profile, and optimize performance-critical components at the kernel and system levels.
• Contribute to the architecture and design of core components for AI compute systems.

Required Skills

• 7+ years of industry experience or a PhD with 5+ years of industry experience.
• Strong understanding of computer architecture, data structures, and system software.
• Proficiency in C/C++ and Python development within a Linux environment.
• Experience implementing algorithms for specialized hardware (FPGA, DSP, GPU, AI accelerators).
• Hands-on experience with CUDA libraries.
• Experience implementing ML operators including GEMMs, convolutions, BLAS, and SIMD operators.
• Experience with embedded SIMD vector processors such as Tensilica.
• MS in Computer Engineering, Math, Physics, or a related field.

Preferred Skills

• Experience with ML frameworks like TensorFlow or PyTorch.
• Experience with ML compilers and algorithms such as MLIR, LLVM, TVM, or Glow.
• Prior experience in a startup, small team, or incubation environment.

Key Skills

Education

MS in Computer Engineering