Principal Software Developer

Responsibilities

• Own ROCm’s end-to-end validation architecture across functional, workload, performance, stress, stability, scale-out, and system-level test layers.
• Define release-qualification gates, exit criteria, coverage standards, performance baselines, stability targets, scale targets, and RAS criteria.
• Architect distributed test runners, CI fleets, hardware-lab orchestration, result data lakes, flaky-test detection, bisection automation, and developer pre-submit pipelines.
• Establish GitHub-based quality workflows, PR gating policies, required checks, code-coverage standards, bug-bash cadences, and issue-management practices.
• Lead root-cause analysis of complex multi-component, multi-node, hardware, firmware, and customer escalations.
• Drive validation of multi-GPU server nodes, PCIe, Infinity Fabric, xGMI, BMC/IPMI, thermal and power behavior, firmware interactions, and Ethernet/InfiniBand/UALink fabrics.
• Lead AI/ML and HPC workload validation for training, inference, recommender systems, scientific kernels, and benchmark suites.
• Mentor Senior and Staff validation engineers, SDETs, and SQA leads through technical reviews and written guidance.
• Influence validation roadmaps for next-generation Instinct GPUs and represent ROCm validation in customer, OEM, and open-source engagements.

Requirements

• BS, MS, or PhD in Computer Science, Computer Engineering, or a related discipline, or equivalent demonstrated experience.
• Software engineering experience in validation, SDET, or quality engineering, including leadership of complex systems validation.
• Expert Python skills for test automation and infrastructure and strong C++ skills for debugging and production-code extensions.
• Deep expertise in at least two relevant areas, including GPU software stacks, AI/ML frameworks, HPC runtimes and communication libraries, Linux kernel or drivers, accelerator firmware, or distributed systems.
• Experience validating multi-GPU, multi-node server platforms through stress, soak, fault-injection, and RAS testing.
• Experience defining release-qualification programs for hyperscalers, OEMs, or Tier-1 customers.
• Experience contributing to validation, CI, or test infrastructure for ROCm, PyTorch, LLVM, Triton, vLLM, or similar open-source projects.
• Experience with agentic AI workflows, automated testing, AI-driven debugging, MCP, and RAG-based engineering solutions.
• Experience validating or operating large-scale GPU clusters of 256 or more GPUs, including fabric bring-up, health monitoring, and diagnostics.
• Familiarity with AI training, inference, HPC benchmark methodologies, performance validation, profiling tools, hardware-lab automation, and pre-silicon or first-silicon accelerator bring-up.

Benefits

• Hybrid role located in San Jose, California.
• AMD benefits are offered; details are provided through AMD’s benefits overview.