AI Research Scientist

AMD is a technology company advancing processors, AI, and computing technologies. The AI Research Scientist will lead strategic research and software engineering efforts in AI, AI for Science, high-performance computing, and energy-efficient computing, developing technologies and prototypes from concept through deployment. The role also involves evaluating workloads, guiding future hardware and software capabilities, and collaborating with internal teams and external research and industry partners.

Responsibilities:

• Work with AMD’s AI, architecture, software, systems, and research specialists to identify and develop technologies that influence future products and platforms across AI, HPC, AI for Science, and energy-efficient computing
• Apply a data minded and scientific approach to workload characterization, model and system evaluation, performance analysis, and identification of architectural and software opportunities
• Stay informed of AI model, agentic AI, scientific computing, software, system, and hardware trends and innovations, especially pertaining to model training, inference, distributed computing, algorithms, memory systems, interconnects, and emerging architectures
• Design and develop new groundbreaking AMD technologies spanning AI model training and inference, agentic AI, AI for Science, scientific computing, heterogeneous systems, and energy-efficient computing
• Participating in the definition, evaluation, and early enablement of important AI, HPC, and AI for Science workloads on new AMD GPUs, CPUs, systems, and software platforms, including using application insights to help guide future hardware and software capabilities
• Debugging/fix existing issues and research alternative, more efficient ways to train, deploy, scale, and execute AI, scientific, and HPC workloads, with particular emphasis on performance, scalability, energy efficiency, and total cost of computation
• Develop technical relationships with peers and partners, including leading collaborations with AI labs, national laboratories, universities, supercomputing centers, open-source communities, strategic customers, and other external organizations