GPU Performance Scientist: CUDA, Triton & Transformers

Luma; Greater London; On-site; GBP 100,000 - 150,000; Full time

Job summary

Luma is seeking a world-class GPU optimization engineer to accelerate multimodal training and deployment. You’ll profile and optimize CUDA, Triton, and PyTorch kernels, delivering maximum throughput with minimal latency on GPU, CPU, and accelerators.

You’ll own fused kernels, tensor cores usage, and distributed multi-node deployment, and you will build monitoring tools, applying cutting-edge transformer optimization techniques.

Qualifications

• Expert-level Triton/CUDA programming and GPU optimization.
• Strong PyTorch skills, including kernel development and custom operations.
• Proficiency with profiling tools (NVIDIA Nsight, torch profiler, custom tooling).
• Deep understanding of transformer architectures and attention mechanisms.

Responsibilities

• Profile and optimize GPU/CPU/accelerator code for maximum utilization and minimal latency.
• Write high-performance PyTorch, Triton, and CUDA, dropping to custom operations when needed.
• Develop fused kernels and leverage tensor cores and modern hardware features across platforms.
• Optimize model architectures and implementations for distributed multi-node production deployment.
• Build performance monitoring and analysis tools and automation.
• Research and implement cutting-edge optimization techniques for transformer models.

Skills

Triton/CUDA programming

GPU optimization

PyTorch skills

Profiling tools

Tools

NVIDIA Nsight

Torch Profiler

Custom tooling

Job description

Luma is seeking a world-class GPU optimization engineer to accelerate multimodal training and deployment. You’ll profile and optimize CUDA, Triton, and PyTorch kernels, delivering maximum throughput with minimal latency on GPU, CPU, and accelerators.

You’ll own fused kernels, tensor cores usage, and distributed multi-node deployment, and you will build monitoring tools, applying cutting-edge transformer optimization techniques.