Technical Marketing Engineer - AI Workloads & ROCm Software/Tools

About the position

At AMD, the customer experience is paramount, and delivering exceptional performance is key to our success. As a Lead Technical Marketing Engineer within the Data Center GPU & Accelerated Processing group, you will be laser-focused on making it easy for customers to achieve the peak performance for AI model workloads running on Instinct GPUs. You'll be at the helm of our efforts to ensure that every customer maximizes the potential of their AI applications in areas such as large language models (LLMs), computer vision, and industry benchmarks. Your expertise will enable customers to understand the true power of our GPUs, driving the adoption and satisfaction that are hallmarks of our customer-centric philosophy.

Responsibilities

• Partner with AMD's AI software engineering team to develop customer facing performance-focused content, including optimization guides, benchmarking results, and performance tuning advice.
• Cultivate comprehensive knowledge of AMD's ROCm software and industry-standard ML frameworks to produce technical materials that drive GPU performance optimization.
• Curate a library of performance optimization documentation, including detailed use-case guides and performance troubleshooting manuals.
• Keep abreast of the latest GPU technology and software developments, ensuring our performance content is aligned with the newest enhancements and capabilities.
• Create and maintain performance-centric code examples, detailed optimization recipes, and ready-to-deploy containers that leverage AMD's ROCm stack for peak AI workload efficiency.
• Engage with technical experts to validate documentation against real-world performance scenarios and benchmarking standards.
• Develop workload self-tuning guides with inputs from engineering for end-customer consumption.
• Proactively gather and integrate feedback from technical experts to ensure documentation meets real-world application and customer requirements.
• Mentor and guide the rest of the organization on general performance optimization & testing methods.
• Lead by example to improve the team's technical documentation capabilities, mentoring colleagues to achieve documentation excellence from initial development through to release.

Requirements

• 5+ years of experience in running and optimizing AI workloads with GPU or AI accelerators at scale.
• Advanced programming skills in Python & C/C++, adhering to the highest standards of software design practices.
• Expertise in developing performance-centric technical documentation, with proficiency in Read the Docs, MarkDown, and Jupyter Notebooks.
• Proven track record of writing both explanatory and procedural content which simplifies the complexity of GPU performance for technical audiences.
• Well-versed with the suite of Microsoft Office, Adobe Acrobat, and other utilities pivotal in crafting and distributing customer documentation and training resources.
• Hands-on experience with GPUs, FPGAs, or other machine learning accelerators, including in-depth knowledge of performance-critical APIs and tools like HIP, CUDA, ROCm, or OpenCL.
• History of engagement with customers in high-tech sectors, particularly within hyperscale datacenter or high-performance computing (HPC) landscapes.

Nice-to-haves

• MS or PhD in Computer Science, Computer Engineering, or Electrical Engineering.

Benefits

• AMD benefits at a glance.