This list contains only the countries for which job offers have been published in the selected language (e.g., in the French version, only job offers written in French are displayed, and in the English version, only those in English).
AMD is looking for a Senior Software Engineer for the Developer Tools team, where you will contribute to the development and optimization of GPU profiling tools for HPC, ML and AI workloads. You will work alongside a team of skilled engineers, collaborating to enhance the profiling tools' capabilities and support broader strategic projects, including participating in new ASIC and hardware bring-ups.
Job Responsibility:
Collaborate with architecture, driver, and runtime teams to enable profiling for next-generation AMD GPUs involved in Pre Silicon and Post Silicon activities
Design, develop, and maintain ROCprofiler-SDK components for GPU performance profiling
Implement new features and APIs to enhance profiling capabilities for AI and HPC workloads
Optimize profiling tools for accuracy, scalability, and minimal overhead
Debug and resolve issues in profiling workflows and improve tool reliability
Stay current with GPU architecture advancements and integrate them into ROCprofiler-SDK
Contribute to documentation and developer resources for ROCprofiler-SDK
Requirements:
Strong proficiency in C++ and object-oriented programming
Experience with performance analysis tools or profiling frameworks
Understanding of GPU architecture and system-level performance concepts
Experience with multithreading and concurrency in modern C++
Familiarity with ROCm ecosystem and tools is highly desirable
Experience with Git-based workflows and debugging tools
Familiarity with GPU programming models (HIP, OpenCL, or CUDA) is a plus
Strong problem-solving skills and ability to work independently and in a team
Knowledge of Linux development environments
Windows experience is a plus
Bachelor’s or Master’s degree in Computer/Software Engineering, Computer Science, or related technical discipline
Nice to have:
Familiarity with GPU programming models (HIP, OpenCL, or CUDA)