The role is for a senior software engineer at AMD, focusing on enhancing the performance of AI workloads on GPUs. The engineer will work within a core team of specialists to influence the ROCm ecosystem and improve how models train and run on AMD hardware. This position requires a blend of deep systems performance expertise and modern AI knowledge. The engineer will tackle complex technical challenges and mentor team members as AMD scales its AI software strategy.

• Expert-level proficiency in modern C++ and design of large, performance-critical systems.
• Strong understanding of GPU architecture, memory hierarchy, and kernel optimization using HIP/CUDA.
• Hands-on experience with large-scale C++/HIP/CUDA codebases, including ROCm and the CUDA ecosystem.
• Deep understanding of transformers and the full model lifecycle in AI.
• Substantial professional experience in software development within performance-critical environments.

• Own the AI software stack and establish best practices for performance from GPU kernels to distributed systems.
• Accelerate foundation models and agents by improving training and inference for LLMs and autonomous AI workloads.
• Co-design hardware and software by partnering on the full lifecycle from GPU architecture input to software for new accelerators.
• Diagnose bottlenecks in multi-GPU, distributed settings using profilers.
• Mentor team members and influence direction across teams to scale AMD's AI software strategy.

• AMD provides a competitive 'Total Rewards' package that focuses on financial growth, health, and work-life balance.