Lead AI/ML Validation Engineer
AMD · Bengaluru · Posted 2026-07-21
Tech stack: C++, GitHub Actions, Jenkins, Linux, Python
Apply on the company site · Get a referral for this role
AMD salary & ratings · AMD interview process · More live openings
About the role
Role Overview We are seeking an experienced SMTS Software Quality Engineer to join the HIP Runtime Quality Engineering team within the ROCm ecosystem. In this role, you will focus on defect triage, root‑cause analysis, and scalable test automation for the HIP runtime, ensuring robustness, portability, and performance across AMD and NVIDIA GPU backends.
Responsibilities:
- Defect Triage & Root Cause Analysis
- Own end-to-end triage of HIP runtime failures observed in CI, nightly, and release testing pipelines.
- Analyze failures across HIP runtime APIs, including memory management, streams, events, kernels, graphs, and device management.
- Perform log analysis, repro reduction, and backtrace debugging across Linux and Windows environments.
- Correlate failures across: ROCr system runtime HIP runtime Driver / firmware interfaces Compiler (hipcc / clang) interactions
- ROCr system runtime
- HIP runtime
- Driver / firmware interfaces
- Compiler (hipcc / clang) interactions
- File high‑quality bug reports with clear reproduction steps, root cause hypotheses, and impact assessment.
- Test Automation & Framework Development
- Design, enhance, and maintain automated test frameworks for HIP runtime validation using existing HIP test infrastructure (Catch2-based and custom harnesses).
- Expand functional, negative, stress, and regression coverage for: HIP Runtime APIs Graph APIs Memory copy variants (1D/2D/3D, async/sync) Multi-GPU and peer-to-peer use cases
- HIP Runtime APIs
- Graph APIs
- Memory copy variants (1D/2D/3D, async/sync)
- Multi-GPU and peer-to-peer use cases
- Integrate tests into CI/CD pipelines with actionable failure reporting.
- Develop Python, C++, and shell-based automation for test execution, result aggregation, and trend analysis.
- Drive test flakiness reduction and improve signal-to-noise ratio in large-scale CI.
- Cross-Team Collaboration
- Partner with HIP runtime developers to: Review test gaps for new features and APIs Validate bug fixes and performance-sensitive code paths
- Review test gaps for new features and APIs
- Validate bug fixes and performance-sensitive code paths
- Collaborate with: Compiler and language runtime teams ROCr system runtime teams Release engineering and CI infrastructure teams
- Compiler and language runtime teams
- ROCr system runtime teams
- Release engineering and CI infrastructure teams
- Contribute test fixes and enhancements upstream in open-source HIP/ROCm repositories.
- Quality Metrics & Continuous Improvement
- Define and track quality metrics such as failure escape rate, test coverage, and mean-time-to-resolution (MTTR).
- Identify systemic issues and propose data-driven quality improvements.
- Participate in release readiness assessments for HIP runtime drops.
Qualifications:
- Technical Skills
- Strong experience in software quality engineering, test automation, or SDET roles.
- Solid proficiency in C++ with debugging experience (gdb/lldb).
- Working knowledge of HIP or CUDA runtime concepts, including kernels, streams, memory models, and device execution.
- Experience with Linux-based development and debugging.
- Familiarity with CI systems (Jenkins, GitHub Actions, GitLab CI, or similar).
- Experience writing and maintaining automated tests for low-level systems or runtime libraries.
- Debugging & Analysis Skills
- Ability to reason about runtime abstractions, concurrency, and asynchronous execution.
- Strong skills in failure reproduction, minimization, and log-driven debugging.
- Experience triaging issues across hardware–software boundaries is a plus.
- Experience with GPU programming models (HIP, CUDA, OpenCL).
- Familiarity with ROCm stack components (HIP, ROCr, compilers, drivers).
- Experience testing graph APIs, async execution models, or multi-GPU systems.
- Prior contributions to open-source system software or test frameworks.
- Knowledge of Python-based test orchestration and reporting tools.
- Knowledge and application of AI tools in validation and testing.
- What Success Looks Like
- Faster, higher-quality issue triage with actionable root cause insights.
- Improved automation coverage for HIP runtime APIs and edge cases.
- Reduced CI noise and faster feedback for developers.
- Strong collaboration with runtime and compiler engineers to ship stable, high-quality HIP releases.
Qualifications
- Strong experience in software quality engineering, test automation, or SDET roles.
- Solid proficiency in C++ with debugging experience (gdb/lldb).
- Working knowledge of HIP or CUDA runtime concepts, including kernels, streams, memory models, and device execution.
- Experience with Linux-based development and debugging.
- Familiarity with CI systems (Jenkins, GitHub Actions, GitLab CI, or similar).
- Experience writing and maintaining automated tests for low-level systems or runtime libraries.
- Debugging & Analysis Skills
- Ability to reason about runtime abstractions, concurrency, and asynchronous execution.
- Strong skills in failure reproduction, minimization, and log-driven debugging.
- Experience triaging issues across hardware–software boundaries is a plus.
- Experience with GPU programming models (HIP, CUDA, OpenCL).
- Familiarity with ROCm stack components (HIP, ROCr, compilers, drivers).
- Experience testing graph APIs, async execution models, or multi-GPU systems.
- Prior contributions to open-source system software or test frameworks.
- Knowledge of Python-based test orchestration and reporting tools.
- Knowledge and application of AI tools in validation and testing.
- Faster, higher-quality issue triage with actionable root cause insights.
- Improved automation coverage for HIP runtime APIs and edge cases.
- Reduced CI noise and faster feedback for developers.
- Strong collaboration with runtime and compiler engineers to ship stable, high-quality HIP releases.
Responsibilities
- Defect Triage & Root Cause Analysis
- Own end-to-end triage of HIP runtime failures observed in CI, nightly, and release testing pipelines.
- Analyze failures across HIP runtime APIs, including memory management, streams, events, kernels, graphs, and device management.
- Perform log analysis, repro reduction, and backtrace debugging across Linux and Windows environments.
- Correlate failures across: ROCr system runtime HIP runtime Driver / firmware interfaces Compiler (hipcc / clang) interactions
- ROCr system runtime HIP runtime
- Driver / firmware interfaces
- Compiler (hipcc / clang) interactions
- File high‑quality bug reports with clear reproduction steps, root cause hypotheses, and impact assessment.
- Test Automation & Framework Development
- Design, enhance, and maintain automated test frameworks for HIP runtime validation using existing HIP test infrastructure (Catch2-based and custom harnesses).
- Expand functional, negative, stress, and regression coverage for: HIP Runtime APIs Graph APIs Memory copy variants (1D/2D/3D, async/sync) Multi-GPU and peer-to-peer use cases
- HIP Runtime APIs Graph APIs
- Memory copy variants (1D/2D/3D, async/sync)
- Multi-GPU and peer-to-peer use cases
- Integrate tests into CI/CD pipelines with actionable failure reporting.
- Develop Python, C++, and shell-based automation for test execution, result aggregation, and trend analysis.
- Drive test flakiness reduction and improve signal-to-noise ratio in large-scale CI.
- Cross-Team Collaboration Partner with HIP runtime developers to: Review test gaps for new features and APIs Validate bug fixes and performance-sensitive code paths
- Review test gaps for new features and APIs
- Validate bug fixes and performance-sensitive code paths
- Collaborate with: Compiler and language runtime teams ROCr system runtime teams Release engineering and CI infrastructure teams
- Compiler and language runtime teams
- ROCr system runtime teams
- Release engineering and CI infrastructure teams
- Contribute test fixes and enhancements upstream in open-source HIP/ROCm repositories.
- Quality Metrics & Continuous Improvement
- Define and track quality metrics such as failure escape rate, test coverage, and mean-time-to-resolution (MTTR).
- Identify systemic issues and propose data-driven quality improvements.
- Participate in release readiness assessments for HIP runtime drops.