Senior Failure Analysis Engineer - Test Development

WHAT YOU DO AT AMD CHANGES EVERYTHING

At AMD, our mission is to build great products that accelerate next-generation computing experiences—from AI and data centers, to PCs, gaming and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity and a shared passion to create something extraordinary. When you join AMD, you’ll discover the real differentiator is our culture.

We push the limits of innovation to solve the world’s most important challenges—striving for execution excellence, while being direct, humble, collaborative, and inclusive of diverse perspectives. Join us as we shape the future of AI and beyond.Together, we advance your career.

The ROLE:

The Quality Engineering team is looking for an experienced Senior Failure Analysis Engineer - Test Development to create advanced test methods that surface elusive failures in GPU accelerator platforms. This role is centered on designing custom execution flows that go beyond standard validation, using stress-based scenarios, VPOD environments, AI/ML workloads, and adaptive test logic to make hard-to-capture issues observable and actionable.

The engineer will expand failure analysis capability across lab, factory, and customer-return cases by building test content that improves repeatability, shortens debug cycles, and increases confidence in root cause findings. They will also help shape intelligent test systems that use internal engineering knowledge and live model inference to guide execution decisions in real time. Working across FA, validation, firmware, diagnostics, and data teams, this person will help convert unclear symptoms into testable conditions that accelerate resolution.

THE PERSON:

The ideal candidate is inventive, methodical, and technically versatile, with a strong instinct for designing experiments that reveal behavior hidden under normal test conditions. They are comfortable navigating hardware, firmware, software, and system-level interactions, and know how to choose the right levers—environment, timing, workload composition, instrumentation, or automation—to provoke meaningful behavior. They are effective in VPOD-based test environments, capable of using model-driven compute activity as part of system stimulation, and confident building AI-enabled workflows that draw from team-specific knowledge during execution.

Just as importantly, they can turn messy observations into disciplined experiments, communicate clearly across teams, and document approaches in a way others canreuse.

KEY RESPONSIBILITIES:

• Architect targeted test methods for hard-to-capture platform behaviors across GPU, server, and rack-scale environments.

• Invent new workload patterns, sequencing approaches, and stress combinations that reveal conditions not covered by conventional diagnostics.

• Build and maintain VPOD-based environments that support scalable experimentation, long-duration execution, and controlled reproduction studies.

• Use inference and training activity as systemstimulito probe platform limits, timing sensitivities, and failure-prone operating regions.

• Develop automation, scripting, and orchestration tools to launch workloads,monitor execution, collect logs, and analyze results at scale across Windows and Linux environments.

• Interpret telemetry, logs, and observed signatures to refine experiments, isolate trigger conditions, and improve confidence in reproduced behavior.

• Create AI-enabled execution flows that use internal FA knowledge and live inference to guide test branching, detect emerging patterns, and support faster triage decisions.

• Partner closely with FA, validation, diagnostics, firmware, and manufacturing teams to translate vague symptoms or sporadic field issues into targeted and repeatable test content.

• Document workload intent, test methods, reproduction conditions, and findings clearly so they can be reused across teams and incorporated into future FA workflows.

• Drive continuous improvement of test development methods, workload libraries, andfailurereproduction strategies to expand FA coverage and reduce time to root cause.

PREFERRED EXPERIENCE:

• Proven track recordof…