ML Kernel Development and Optimization Engineer on GPU

About the role

Join AMD's GPU Software team to design, develop, and optimize innovative GPU features that power next‑generation AI and high‑performance computing workloads. You will work across software, machine learning, math libraries, and GPU architecture teams to deliver high‑performance solutions for core workloads such as Convolution, GEMM, and Attention.

Key responsibilities

• Design and develop high‑performance GPU kernels for machine learning and Convolution workloads using HIP.
• Collaborate with GPU architects to explore, evaluate, and leverage emerging hardware features.
• Analyse, profile, and optimise GPU kernel performance, identifying opportunities for continuous improvement.
• Create technical documentation, share best practices, and support the adoption of new capabilities within ROCm.
• Partner with engineers, researchers, customers, and ecosystem partners to enhance ROCm applications, libraries, tools, and hardware.

Required profile

• Collaborative and curious engineer passionate about GPU computing, machine learning, and performance optimisation.
• Strong communication and collaboration skills, able to work effectively across multidisciplinary teams.

Required skills

• Experience with GPU architecture, performance analysis, algorithm development, and machine‑learning workloads.
• Hands‑on experience developing and optimising GPU software using HIP, CUDA, or similar technologies.
• Understanding of GPU runtimes, compilers, machine‑learning frameworks, HPC libraries, and performance‑optimisation techniques.
• Knowledge of software‑engineering best practices, including testing, profiling, debugging, version control, and documentation.