PREFERRED EXPERIENCE:

• Understanding of GPU architectures and low-level optimization techniques, memory hierarchy, instruction scheduling, and performance tradeoffs
• Strong background developing applications and libraries in C++, especially high-performance computing and/or scientific software
• GPU software development using HIP, CUDA, or OpenCL
• In-depth knowledge of best-practices in software development, including testing, profiling, debugging, documentation, version control, issue tracking, and planning
• Experience with C++ and Python programming
• Experience in PyTorch, Tensorflow or MXNet
• Experience in high-performance computing (HPC)
• Strong written and spoken English language skills (our company language)

WHAT YOU DO AT AMD CHANGES EVERYTHING

We care deeply about transforming lives with AMD technology to enrich our industry, our communities, and the world. Our mission is to build great products that accelerate next-generation computing experiences – the building blocks for the data center, artificial intelligence, PCs, gaming and embedded. Underpinning our mission is the AMD culture. We push the limits of innovation to solve the world’s most important challenges. We strive for execution excellence while being direct, humble, collaborative, and inclusive of diverse perspectives.
AMD together we advance_
Responsibilities:

THE ROLE:

AI Performance Engineers focus on optimizing machine learning models, profiling, bottleneck analysis, and optimal mapping to GPU.
As an AI Performance Engineer, you will analyze and explore recent ML models, understand their compute and memory requirements, and optimize them on various compute hardware for both inference and training. In addition you would do profiling and analyzing various workloads on current hardware and coming up with new ways to improve their performance.

KEY RESPONSIBILITIES:

• Benchmark, analyze, and optimize performance of key machine learning applications, on single and multi-GPU systems
• Design, implement, and test GPU kernels and algorithms for tensor operations like matrix multiplication and convolutions used in a variety of high-performance machine learning libraries and frameworks
• Deliver high-quality code and documentation following best practices for open-source software development