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).
In this role, you will provide quality and DevOps engineering support for AMD's GPU-accelerated ML libraries powering AI, LLM, and deep learning applications. You will be responsible for developing and executing comprehensive CI/CD pipelines, automated test suites, ETL data systems, and observability dashboards that validate library correctness across GPU architectures. You will work closely with library developers, platform teams, and program management to ensure stability, reliability, and performance via both automated pipelines and hands-on testing.
Job Responsibility
Test Automation Development: Design, implement, and maintain automated test suites using gtest for an open-source, C++-based library
CI/CD Integration: Integrate test automation frameworks into the GitHub Actions pipeline, ensuring seamless execution of tests and rapid feedback for developers
Tooling: Drive the scaling out of our common platform to collect and visualize Quality metrics, taking a data-driven approach to surfacing engineering health
Bug Detection & Reporting: Identify, isolate, and report defects found during testing and work with developers to prioritize and resolve issues
Continuous Improvement: Continuously improving the test infrastructure and methodologies, proposing tools or techniques that can improve the testability of the codebase
Collaboration & Documentation: Work with cross-functional teams, document test results, and assist in creating user-friendly reports that communicate the quality status of the project
Test Planning: Collaborate with developers and product teams to define test strategies, test cases, and acceptance criteria for new features and enhancements in the library
Code Coverage: Develop and analyze solutions, identify gaps, and drive improvements in both test coverage and quality
Debugging & Root Cause Analysis: Lead complex debugging across the stack (kernel → library → framework), driving systematic resolution and preventing recurring issues
Requirements
B.Tech/M.Tech in Computer Science, Electrical/Electronic Engineering, or related field
10+ years of experience in DevOps, CI/CD, or infrastructure engineering
Proven experience as an SDET or in a similar role with a focus on C++ development and testing
Strong experience with GoogleTest (gtest) for unit and integration testing in C++
Hands-on experience with Jenkins or GitHub Actions for automating test execution and integrating tests into CI/CD pipelines
In-depth knowledge of software testing methodologies, frameworks, and tools for automated testing
Experience working in Windows and Linux-based environments for development and testing
Familiarity with Docker or other containerization for managing test environments and consistent test execution
Familiarity with version control systems such as Git, and with development tools and practices common in open-source communities
Familiarity with building/deploying web apps for the purpose of dashboarding and/or visualization
Excellent problem-solving skills and a proactive approach to testing and debugging
Strong written and verbal communication skills, with the ability to collaborate effectively with both technical and non-technical teams
Experience leveraging agentic AI to automate multi-step engineering workflows such as CI orchestration, test generation, or incident remediation
Nice to have
Experience with ROCm/CUDA, HIP, or GPU compute stacks
Proficiency in C++, including memory management, data structures, and algorithms
Familiarity with pre-silicon simulation environments (functional models, emulators)
Experience with GPU kernel compilation workflows and binary artifact management
Knowledge of JIRA/Confluence integration for traceability and automation
Experience with observability patterns (metrics, alerting, SLA monitoring)
Familiarity with HSA runtime, GPU driver stacks (KMD/UMD), or DTIF environments