Overview
AI Performance Architect, Training and Inference - Santa Clara, CA area
This role is with AMD. Base pay range is provided below; actual pay is based on skills and experience. Talk with your recruiter to learn more.
Base pay range
$226,400.00/yr - $339,600.00/yr
The Role
Join a world-class team enabling software for world-class datacenters and the most powerful supercomputers. AMD is seeking talented and motivated AI Software Engineers to push the boundaries of efficiency and performance, optimizing the software ecosystem for the next generation of GPU accelerators. You will work to combine the newest hardware with the industry’s latest applications, libraries, frameworks, and SDKs to solve complex AI challenges. Minimum 7 years of experience required.
Responsibilities
- Enable DL models, libraries, and applications for Instinct GPUs in on-prem and cloud environments.
- Analyze and optimize the performance of AI software and understand hardware bottlenecks to hit near roofline performance.
- Collaborate with a team of Software Engineers to deliver high-performance AI software products.
Key Qualifications
- Strong programming skills in C++ and Python.
- Experience with at least one major DL framework (e.g., PyTorch or TensorFlow) in inference, fine-tuning, and/or training.
- MS with related experience or PhD in Computer Science/Engineering or related field.
- Experience developing software and system-level performance optimizations with solid GPU architecture understanding (a plus).
- Experience with open-source software development and contributing to communities (a plus).
- Publications in reputed peer-reviewed ML conferences/journals (a plus).
- Excellent analytical and problem-solving skills for root-causing and addressing performance issues.
- Ability to work independently and within a team; self-motivated with a collaborative mindset.
- Willingness to learn tools and methods to improve AMD software quality and timeliness.
Preferred Experience
- Expertise in profiling tools across the AI software stack (TorchProfiler, ROCm Profiler, VTune, Nsight).
- Experience implementing and optimizing parallel methods on GPUs (NCCL/RCCL, OpenMP, MPI).
- Performance analysis skills for CPU and GPU.
- Experience with Singularity, Docker, and/or Kubernetes.
- Ability to communicate status and project insights clearly to leadership.
Location
Santa Clara, CA area
Benefits
Benefits offered are described: AMD benefits at a glance. AMD is an equal opportunity, inclusive employer and will consider all applicants without regard to protected characteristics.
#J-18808-Ljbffr