Principal Software Engineer, SDK & Lowering Stack

Job Title

Principal Software Engineer, SDK & Lowering Stack

Role Summary

Lead design and implementation of the kernel authoring SDK and lowering toolchain targeting d-Matrix's Corsair dataflow architecture. You will work across software and hardware teams to define APIs, build developer tools, and deliver production-quality components of the ML runtime stack.

Location: Santa Clara, CA (headquarters); open to candidates elsewhere in the United States or Canada.

Experience Level

Level: Senior. Typical experience expected: 8+ years in industry (see Education Requirements for degree-and-years combinations).

Responsibilities

Primary responsibilities include:

• Architect and implement the kernel authoring SDK for the Corsair dataflow architecture.
• Design and expose domain-specific, production-grade APIs and SDK boundaries for kernel and compiler teams.
• Develop tooling for kernel authoring and for modeling lowering flows.
• Establish and drive software engineering best practices: CI/CD, automated testing, and rigorous code review.
• Mentor and lead engineers, promoting technical excellence and accountability.
• Analyze hardware–software interactions and propose systems-level solutions.
• Manage end-to-end delivery: requirements gathering, design documentation, implementation, debugging, and deployment.
• Scale software deliverables under tight schedules.

Requirements

Must-have skills and experience:

• Proficient in C/C++ and Python development on Linux using standard build and debugging tools.
• Proven experience establishing CI/CD pipelines, automated testing, and enforcing code-review processes.
• Strong knowledge of computer architecture, data structures, system software, and machine learning fundamentals.
• Experience designing high-quality, scalable APIs and SDKs with a systems engineering mindset.
• Demonstrated ownership, leadership, and ability to deliver complex software projects.

Nice-to-have / preferred:

• Experience writing and optimizing high-performance kernels for GPUs, TPUs, or custom ML accelerators.
• Background in High-Performance Computing and large-scale distributed workload optimization.
• Experience with ML compilers and intermediate representations (e.g., MLIR, LLVM, TVM, Glow).
• Familiarity with AI-assisted coding tools.
• Prior startup, small-team, or cloud/AI compute company experience.

Education Requirements

MS in Computer Engineering, Mathematics, Physics, or a related field plus 8 years of industry experience, or PhD in those fields plus 4 years of industry experience.

About the Company

Company: d-Matrix

Headquarters: Santa Clara, California, United States

d-Matrix is a Santa Clara–based startup developing highly programmable in-memory computing architectures and accompanying software to accelerate generative AI and other AI workloads, focusing on hardware-software co-design for cloud and edge applications.