Principal AI Performance Engineer - GPU Kernels
Advanced Micro Devices Näytä kaikki työpaikat
- Helsinki
- Vakituinen
- Täyspäiväinen
- Lead the identification and implementation of improvements to machine-learning kernels for AMD GPUs, with a focus on performance and power efficiency
- Define technical strategy and best practices for kernel development within the team
- Stay informed about software and hardware trends, particularly in GPU architecture and ML algorithms, and translate insights into actionable roadmap items
- Improve development workflows and CI infrastructure to enable faster and more reliable delivery
- Architect and develop new, innovative GPU and ML technologies
- Debug and resolve complex issues, and research alternative, more efficient implementations
- Mentor junior and mid-level engineers, raising the technical bar across the team
- Build and maintain strong technical relationships with internal teams and external partners, influencing upstream roadmaps and priorities
- Deep understanding of modern GPU architectures and micro-architectural performance characteristics
- 7+ years of GPU software development experience using HIP, CUDA, or OpenCL
- Experience working directly with Hardware ISA is a big plus
- 8+ years of system-level programming experience in C++ and/or Python (C++17 or later preferred)
- Proven track record of leading technical projects from design through delivery
- Experience with GPU profiling, debugging, benchmarking, and performance analysis tools
- Background in high-performance computing (HPC) or other performance-critical systems
- Familiarity with modern ML frameworks such as PyTorch, vLLM and SGLang
- Experience with tile-based programming models and frameworks (e.g., Triton/Gluon, CUTLASS, CK)
- Experience with GPU compiler toolchains (e.g., LLVM) and intermediate representations (e.g., MLIR, LLVM IR, Triton IR) is a strong plus
- Experience mentoring engineers and driving technical excellence within a team
- Strong written and verbal English communication skills
- Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent. PhD is a plus.