Principal Software Engineer, Kernels
Develop and optimize software kernels for next-gen AI hardware, mapping computational graphs from ML frameworks to d-Matrix's compute engine architecture. Work across the full SW stack with compilers (MLIR/LLVM/TVM), implement ML operators (GEMMs, convolutions, softmax), and collaborate with hardware and compiler teams. Core tech: C/C++, Python, CUDA, SIMD vector processors.
- Compensation
- $195K–$285K / yr
- Location
- Santa Clara
- Work setup
- Hybrid
- Seniority
- Principal
- Employment
- Not specified
- Experience
- 5+ years
- Category
- Software engineering
- Posted
Skills
Nice to have
- Prior startup, small team, or incubation experience.
- Experience with ML frameworks such as TensorFlow and/orPyTorch.
- Experience working with ML compilers and algorithms, such as MLIR, LLVM, TVM, Glow, etc.
- Experience with a deep learning framework (such as PyTorch or TensorFlow) and ML models for CV, NLP, or recommendation.
- Work experience at a cloud provider or AI compute/subsystem company.
Full description
Join d-Matrix and help build the future of AI inference. We’ve pioneered a memory-centric architecture that brings compute directly into memory, making AI faster, more energy efficient, and accessible on a global scale. From chips and software to rack-scale systems, our team is tackling some of AI infrastructure’s hardest challenges — and shipping breakthrough technology today. If you want to create what comes next and help make infinite inference possible, we’d love to hear from you.
Our culture is built on respect, collaboration, and direct communication. We value humility, we believe our different perspectives lead to better solutions, and we’re looking for people who are driven by execution and energized by hard problems.
Location:
Hybrid, working onsite at our Santa Clara, CA, headquarters 3+ days per week.
The Role: Principal Software Engineer, Kernels
What you will do:
The role requires you to be part of the team that helps productize the SW stack for our AI compute engine. As part of the software team, you will be responsible for the development, enhancement, and maintenance of software kernels for next-generation AI hardware. You possess experience building software kernels for HW architectures. You possess a very strong understanding of various hardware architectures and how to map algorithms to the architecture. You understand how to map computational graphs generated by AI frameworks to the underlying architecture. You have had past experience working across all aspects of the full-stack toolchain and understand the nuances of what it takes to optimize and trade off various aspects of hardware-software co-design. You can build and scale software deliverables in a tight development window. You will work with a team of compiler experts to build out the compiler infrastructure, working closely with other software (ML and systems) and hardware (mixed signal, DSP, and CPU) experts in the company.
What you will bring:
Minimum:
- MS in computer engineering, math, physics, or a related degree with 10+ years of industry experience or a 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.
- Proficient in C/C++ and Python development in Linux environments and using standard development tools.
- Experience implementing algorithms in high-level languages such as C/C++ and Python.
- Experience implementing algorithms for specialized hardware such as FPGAs, DSPs, GPUs, and AI accelerators using libraries such as CUDA, etc.
- Experience in implementing operators commonly used in ML workloads—GEMMs, Convolutions, BLAS, SIMD operators for operations like softmax, layer normalization, pooling, etc.