System-Level ML Runtime & Driver Engineer (GPU/NPU) - ATG

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:

Join AMD's Advanced Technologies Group (ATG) to develop and optimize machine-learning runtimes and device drivers for AMD GPU/NPU platforms. You'll enable ONNX Runtime/Windows ML workloads through DirectML/MIGraphX/ROCm EPs and collaborate with compiler teams on graph optimizations that deliver performance, power, and memory efficiency across computer vision and LLM use cases.

THE PERSON:

The ideal candidate should be passionate about software engineering and possess leadership skills to drive sophisticated issues to resolution. Able to communicate effectively and work optimally with different teams across AMD.

KEY RESPONSIBILITIES:

• Design and implement ML runtime components and device drivers for AMD machine learning workloads on GPU/NPU platforms.
• Integrate and optimize execution providers (ONNX Runtime DML/ROCm EP, MIGraphX); validate accuracy, latency, and initialization overhead on AMD silicon.
• Partner with compiler teams to land MLIR/TVM/IREE/XLA passes and fused kernels; drive vectorization, layout, and operator fusion aligned with hardware capabilities.
• Profile and tune kernels and graphs using system-level tools; publish analyses (roofline, memory hierarchy) and automate perf regression.
• Collaborate with hardware and architecture teams to provide feedback and direction on design decisions based on in-depth performance analysis.
• Support new ASIC bring-ups and cross-stack debugging; build diagnostic tooling, samples, and developer docs for AMD's ML SDKs.
• Collaborate with Microsoft/OEM partners to demonstrate performant Windows ML solutions on Strix Halo; contribute to GDC-facing demos and materials.

PREFERRED EXPERIENCE:

• Strong C/C++ (C++17+); concurrency and multithreaded runtime/driver programming.
• Linux/Windows device driver and runtime development for GPU/NPU/DSP; familiarity with HALs and kernel subsystems.
• Strong background in compute system architecture (GPU, NPU). Solid understanding of both software and hardware design and analysis.
• Experience with ML frameworks like ONNX/PyTorch and at least one compute stack (ROCm/HIP, CUDA, OpenCL, Metal/Vulkan compute).
• Proven profiling & performance tuning of kernels and graph runtimes.
• Execution providers (DML/ROCm/MIGraphX) and Windows ML; optimizing initialization time and fused kernel libraries (e.g., hipDNN/MIGraphX).
• Open-source contributions in ROCm, LLVM/MLIR, TVM, ONNX Runtime.
• ML compiler & graph optimization expertise: DirectML, CoreML/MPS Graph, MLIR/TVM/XLA/IREE, quantization/mixed precision (INT8/FP8); autotuning/auto-scheduling.

ACADEMIC CREDENTIALS:

• Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent

The role is not Visa eligible for sponsorship

#LI-JG1

Benefits offered are described: AMD benefits at a glance.

AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services. AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law. We encourage applications from all qualified candidates and will accommodate applicants' needs under the respective laws throughout all stages of the recruitment and selection process.