3 months ago
Base Salary
$190k - $300k/yr
Responsibilities
- Develop, enhance, maintain, and productize software kernels for ML operators on next-generation AI hardware.
- Implement algorithms and operators including GEMMs, convolutions, softmax, layer normalization, pooling, and activation functions.
- Map algorithms and AI-framework computational graphs onto modern hardware architectures.
- Build SDK solutions that are intuitive for developers to use and support performance analysis.
- Work across the software toolchain and navigate hardware–software co-design trade-offs while delivering scalable, high-quality software.
Requirements
- Master’s or PhD in computer engineering, mathematics, physics, or a related field, plus 5 or more years of industry experience.
- Strong understanding of computer architecture, data structures, system software, and machine learning fundamentals.
- Proficiency in C/C++ and Python development in a Linux environment using standard development tools.
- Experience implementing algorithms in C/C++ and Python.
- Experience targeting specialized hardware such as FPGAs, DSPs, GPUs, and AI accelerators, including libraries such as CUDA.
- Experience implementing common ML workload operators such as GEMMs, convolutions, softmax, layer normalization, and pooling.
- Preferred experience in a startup, small-team, or incubation environment.
- Preferred experience with TensorFlow, PyTorch, ML compilers and algorithms such as MLIR, LLVM, TVM, and Glow, deep learning models for computer vision, NLP, or recommendation, embedded SIMD vector processors such as Tensilica, or cloud and AI compute companies.
Benefits
- Work location is Santa Clara, California headquarters or a regional office, with remote work considered.
About d-Matrix
d-Matrix builds AI inference computing platforms for data centers, combining custom silicon with systems, networking, and software. Its flagship Corsair platform and JetStream fabric focus on low-latency, energy-efficient generative AI inference at scale. Founded in 2019 and headquartered in Santa Clara, California, the privately held company sells hardware with accompanying software to cloud providers and enterprises deploying large AI models.