Quality engineer (auto) for AI inference

грейд
senior
формат
офис Белград
страна
Сербия
язык
Английский B1
опубликована
сегодня

описание

Join d-Matrix and help build the future of AI inference. We’ve pioneered a memory-centric architecture that brings compute directly into memory, making AI faster, more energy efficient, and accessible on a global scale. From chips and software to rack-scale systems, our team is tackling some of AI infrastructure’s hardest challenges — and shipping breakthrough technology today. If you want to create what comes next and help make infinite inference possible, we’d love to hear from you.

задачи

  • Work with QE and Engineering to understand users' needs and agree priorities. Define scope and expected results. Raise dependencies and blockers, and ask for help reaching the right owner.
  • Build and maintain shared automation, test frameworks and harnesses for functional, integration, regression and release testing across supported software and hardware configurations.
  • Turn scripts and prototypes into tools others can use. Provide clear interfaces, configuration, setup instructions, error handling and diagnostics. Review code and test the tools.
  • Integrate tools with existing continuous integration (CI), nightly and release workflows alongside infrastructure owners. Record builds, configurations, commands, inputs and results so another engineer can reproduce failures.
  • Support adoption through documentation, examples and user feedback. Check that colleagues can use tools independently. Measure human effort, reliability and result quality, including review and maintenance costs.
  • Assess coverage, assertions, expected results and negative cases. Distinguish product deficits, environment problems and unreliable tests. Work with developers to verify fixes and provide testing cover while progressing shared automation.
  • Keep up with advances in AI coding agents and workflow tools. Test promising approaches on real QE work, share what you learn and build on the ones that improve results. Use approved tools, check outputs and follow approved data-handling rules.
  • Understand one QE workflow, its users, dependencies and failure modes. Establish an improvement baseline.
  • Deliver a reviewed automation improvement for an agreed need in the existing test environment.
  • Show reliable repeat execution, useful diagnostics and reproducible results against identified builds and configurations.
  • Have intended users set up, run and troubleshoot the tool from its documentation. Act on their feedback.
  • Compare effort, reliability or result quality with the baseline. Hand over operating and maintenance notes.
  • Agree scope, targets and measurement windows with the manager after onboarding.

требования

  • 5+ years of experience
  • Strong software-development skills: write, debug and maintain code, explain design choices, review others' work and investigate failures beyond a test's surface result.
  • Experience building shared test automation, frameworks or engineering tools that others use repeatedly.
  • Evidence of turning prototypes or manual workflows into reliable tools, including configuration, failure handling, documentation, adoption and maintenance.
  • Practical test engineering: requirements, boundary and negative cases, coverage, and failure investigation using code, logs and reproducible experiments.
  • Experience with Linux, version control, code review, automated tests and CI workflows.
  • Regular use of AI coding agents or AI workflows, with examples of what you built, how your approach evolved and how you checked the results.
  • Experience guiding technical work across distributed teams through design and code reviews. Explain tradeoffs, understand users' needs, make progress independently and raise missing decisions promptly.
  • Будет плюсом: Systems software testing in firmware, drivers, compilers, runtimes or distributed systems, including real devices; ML inference, model execution, numerical comparisons, quantization or performance benchmarks; Test frameworks, CI pipelines, containers, reporting or lab automation; Hardware-in-the-loop, multi-card or multi-node configurations, and extended stability or soak testing; Internal engineering products used across teams, or human-reviewed AI workflows used repeatedly

условия

  • Условий в вакансии нет

навыки

похожие вакансии

статьи для QA-инженеров