Senior Software Engineer - GPU Profiling Tools
Overview:
In this role you will help build and optimize GPU profiling tools for HPC and AI workloads at AMD. You will join the Developer Tools team to advance ROCprofiler-SDK and contribute to pre- and post-silicon profiling efforts. You will work with firmware, architecture, driver, and runtime teams to enable high-quality, low-overhead profiling across AMD GPUs. This is a hands-on, impact-driven position that supports AI, HPC, and hardware bring-ups in a fast-paced, cross-functional environment.
Responsibilities:
- Collaborate with firmware, architecture, driver, and runtime teams to enable profiling for next-generation AMD GPUs in pre-silicon and post-silicon environments
- Design, develop, and maintain ROCprofiler-SDK components for GPU performance profiling
- Implement firmware, driver, and runtime interfaces for low-overhead profiling and telemetry features
- Add features and APIs to enhance profiling for AI and HPC workloads
- Optimize profiling tools for accuracy, scalability, and minimal overhead
- Debug and resolve issues across application, runtime, driver, firmware, and hardware layers
- Stay current with GPU architecture advancements and integrate them into ROCprofiler-SDKContribute to documentation and developer resources for ROCprofiler-SDKKey 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 Linux kernel development, kernel drivers, firmware, or low-level system software
- Experience with performance-monitoring technologies such as tracing or hardware counters
- Experience debugging across applications, runtimes, drivers, firmware, and hardware interfaces
- Experience with multithreading and concurrency in modern C++Familiarity with ROCm ecosystem and tools is highly desirable
- Experience with AI-powered developer tools and workflows including code generation, debugging, performance optimization
- Experience with Git-based workflows and debugging tools
- Familiarity with GPU programming models (HIP) is a plus
- Knowledge of Linux development environments; Windows is a plus
- Cross-functional collaborationstrong problem-solvingeffective communicationC++ (OO)GPU profiling tools
- Linux kernel development