Fellow Software Development Engineer, GEMM Optimization

What you'll do

• Architect and develop features in the tensile-lite assembly code generator for GEMM and GEMM+X operations.
• Analyze GPU hardware specifications to align kernel designs with current silicon capabilities.
• Influence future hardware requirements by partnering directly with hardware architects.
• Develop support for new ISA and hardware/software optimization features in the code generator.
• Innovate and implement new algorithms for GEMM and GEMM+X performance.
• Profile and root-cause performance bottlenecks across the software and hardware stack.
• Partner with customers to identify real-world workload requirements and resolve performance issues.

What we're looking for

• Bachelor's, Master's, or PhD in Computer Science, Computer Engineering, or a related field.
• Strong command of GPU computer architecture including compute units, memory hierarchies, and instruction scheduling.
• Deep expertise in GEMM algorithms and their mapping onto GPU hardware.
• Proficiency in C++ and assembly-level GPU programming.
• Working knowledge of Python for tooling and automation.
• Experience optimizing GPU compute kernels, specifically GEMM and attention.
• Direct experience with any GPU compiler backend.
• Experience with AMD GPU architectures (CDNA/RDNA) or comparable competitive architectures.