AMD is seeking a Principal Software Engineer to lead ROCm software validation for compute workloads and server-class systems. This individual-contributor role involves defining validation strategies and building test infrastructure to ensure ROCm is ready for release. The engineer will tackle complex debugging and qualification challenges, directly impacting the quality experienced by customers and the open-source community. The position requires collaboration with various teams to influence product roadmaps and mentor junior engineers.

• Software engineering experience in validation, SDET, or quality engineering.
• Expertise in Python for test automation and strong C++ skills for debugging.
• Deep validation expertise in GPU software stacks, AI/ML frameworks, or HPC runtimes.
• Experience validating multi-GPU, multi-node server platforms.
• BS/MS/PhD in Computer Science, Computer Engineering, or related discipline.

• Own the end-to-end validation architecture for ROCm across multiple GPU generations.
• Define release-qualification gates and exit criteria for ROCm software releases.
• Architect the test infrastructure including distributed test runners and CI fleets.
• Champion modern quality engineering practices and continuous validation.
• Lead complex escalation debug efforts and root-cause multi-component failures.
• Influence the roadmap for next-generation GPUs and server platforms.
• Mentor and elevate validation engineers and SQA leads.
• Represent ROCm validation in strategic customer engagements and open-source initiatives.
• Drive compute workload validation and characterization for various AI/ML frameworks.

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