The Software and AI (SAI) organization is seeking a software development engineer to contribute to the oneDNN project, which is essential for optimizing neural network performance across various AI applications. This role involves working closely with software developers to enhance AI frameworks and workloads for Intel CPUs and GPUs. The position requires expertise in software development, particularly in performance engineering and open-source contributions. The team is focused on delivering high-performance solutions that support Intel's AI strategy.

• Master’s degree or Ph.D. in Mathematics, Physics, Computer Science, or a relevant STEM field.
• 5+ years of experience in C and C++.
• Experience maintaining or contributing to open-source software projects.
• Knowledge in software libraries design and architecture.
• Experience implementing linear algebra algorithms from BLAS, LAPACK, or PyTorch.
• Expertise in performance engineering and software performance optimizations.
• Understanding of floating point arithmetic and numerical stability.
• Experience in software development on Linux.
• Skills in low-level performance optimizations using CUDA, x86 assembly or intrinsics, or OpenCL.

• Design, develop, and maintain new functionality in oneDNN to enhance performance for AI workloads.
• Support software developers in optimizing AI frameworks and workloads for Intel hardware.
• Collaborate with cross-platform ecosystem of AI software developers contributing to oneDNN.
• Engage in performance engineering to ensure high efficiency of neural network applications.

• Intel offers a comprehensive benefits package including competitive pay, stock programs, healthcare coverage, retirement plans, paid time off, parental leave, and programs supporting employee wellbeing and professional development.