Linux Kernel IQE Testing
Integration and Qualification Engineering (IQE) for AMDGPU Linux kernel development — regression testing of kernel patches and firmware on MI300X and NAVI data-center GPUs in BareMetal and SR-IOV environments.
Project Overview
Joined AMD's GPU software team and started in IQE to build deep familiarity with the Linux kernel driver stack before contributing to RAS tool development. Led the migration from manual Bash-based test scripts to a maintainable Pytest framework integrated into Jenkins CI/CD pipelines.
Key Contributions
Performed integration and regression testing of AMDGPU kernel patches and firmware before customer release on MI300X and NAVI GPU families.
Built initial Bash CI automation for test execution, then led migration to Pytest for better maintainability, extensibility, and test coverage reporting.
Designed and executed test plans for BareMetal and SR-IOV deployments on server-class infrastructure with multiple VMs.
Automated validation for RAS, AMD-SMI, ROCm-SMI, and RDC tools — reducing manual regression effort and release risk.
Partnered with KFD/KMD developers, QA, and Cloud Virtualization Support teams to reproduce failures, accelerate debugging, and improve release quality.