Senior Research Engineer (Hardware Accelerators)

About Code Metal Code Metal is redefining code translation for mission-critical industries, helping defense partners move more quickly and reliably from algorithm to silicon. Our platform accelerates deployment of DSP, RF, communications, and embedded signal processing algorithms onto heterogeneous compute targets, including GPUs, FPGAs, ASICs, and edge SoCs. We also support automotive, aerospace, and semiconductor partners deploying complex algorithms onto constrained hardware with speed and rigor. Code Metal is seeking a Research Engineer specializing in programming of hardware accelerators to join our team. In this role, you will collaborate closely with researchers and engineers to design and develop code transpilation and optimization pipelines for various hardware accelerators such as GPUs, NPUs, FPGAs, etc. You will work at the intersection of theory and practice, applying, implementing, and integrating parallel programming and performance analysis related tools into Code Metal products.

Responsibilities Work closely with the research and engineering team to build code transpilation and optimization pipelines for various hardware accelerators. At least one of the source and target languages would be accelerator-specific (such as CUDA, HIP, SYCL). The source language could also be a generic language such as Python or C/C++. Understand the architecture of the target hardware and propose optimization strategies that would accelerate given input code Analyze performance of the input as well as the transpiled code to determine optimization strategies Integrate hardware specific tools (compilers, profilers, libraries, etc.) in the transpilation and optimization pipelines Conduct experiments to evaluate performance of code transpilation and optimization pipelines. Analyze experimental results to improve pipelines further. Collaborate across multidisciplinary teams, contributing to both research initiatives and production-grade solutions. Present research findings in a broader audience group. Required

Qualifications 5+ years of industry experience and an advanced degree (Master’s or Ph.D.) in Computer Science, Mathematics, or a related field, with a focus on software engineering, high-performance computing, and parallel programming languages. Strong programming skills in Python and C/C++. Good understanding of parallel programming languages: CUDA/OpenCL/OpenMP/SYCL. Good understanding of popular hardware devices such as CPUs and GPUs. Understanding of AI-specific hardware accelerators such as NPUs is a plus. Good understanding of common performance analysis practices Demonstrated proficiency in core software development practices, including: version control, code review, testing, continuous integration, build systems, and release pipelines. Combination of strong software engineering skills with a solid background in high-performance computing to build robust and maintainable systems. Ability to translate technical research into practical engineering solutions. Excellent communication skills and a collaborative mindset. Preferred