GPU Kernel Developer - AI Models

As a GPU Kernel Developer specialized in AI Models, your role at AMD involves developing high performance kernels for cutting-edge GPU hardware to support AI/HPC applications. You will collaborate with a team of industry experts to optimize GPU code using a structured approach, analyze performance gaps, and implement necessary optimizations. Your responsibilities will include supporting critical workloads in NLP/LLM, Recommendation, Vision, and Audio, collaborating with various teams to enhance AI training and inference efficiency, and engaging with open-source framework maintainers to integrate code changes. Additionally, you will debug, maintain, and optimize GPU kernels while driving AI operator performance improvements. Key Responsibilities: - Develop high performance GPU kernels for key AI operators on AMD GPUs - Optimize GPU code through structured methodology, profiling, and identifying key optimizations - Support critical workloads in NLP/LLM, Recommendation, Vision, and Audio - Collaborate with various teams to enhance AI training and inference efficiency - Work with open-source framework maintainers to integrate code changes - Debug, maintain, and optimize GPU kernels to improve AI operator performance - Apply software engineering best practices Preferred Experience: - Knowledge of GPU computing platforms like HIP, CUDA, OpenCL, Triton - Experience in optimizing GPU kernels and using profiling tools - Understanding of GPU hardware architecture - Proficiency in C/C++/Python programming with strong software design skills Qualifications Required: - Masters or PhD in Computer Science, Computer Engineering, or related field ,