Founding GPU Engineer — CUDA Performance for HPC

Fuse Energy, LLC; Greater London; On-site; GBP 90,000 - 130,000; Full time

Benefits offered by this job

Competitive salary

Equity sign-on bonus

Biannual bonus

Fully expensed tech

Breakfast and dinner allowance

Job summary

Fuse Energy, LLC seeks an experienced CUDA performance engineer to design, optimize, and deploy high-throughput CUDA kernels across multi-GPU systems. You will profile hardware bottlenecks, develop tooling to monitor energy usage, and collaborate with ML engineers to integrate kernels into scalable training and inference pipelines.

The role requires deep GPU architecture knowledge, strong C++/CUDA skills, and experience with Nsight, NCCL, and MPI.

Qualifications

• 4+ years of experience writing production CUDA code or equivalent
• Deep understanding of GPU architecture (SMs, warps, memory hierarchy)
• Proficiency in C++ and CUDA; Python for tooling/orchestration
• Experience with performance profiling tools (Nsight) and multi-GPU scaling (NCCL, MPI)
• Strong memory optimisation and parallel algorithm design

Responsibilities

• Design, implement, and optimise CUDA kernels for high-throughput, latency-sensitive workloads.
• Profile and tune GPU performance across compute and memory bottlenecks.
• Build tooling to correlate GPU cluster power draw and utilisation with energy pricing and grid signals.
• Optimise multi-GPU and multi-node scaling using NCCL, MPI, or similar libraries.
• Collaborate with ML/systems engineers to integrate kernels into training/inference pipelines.

Skills

CUDA kernel development

GPU architecture knowledge

C++ and CUDA

Python tooling

Performance profiling tools

Multi-GPU scaling

Tools

Nsight Systems

NCCL

MPI

RDMA/InfiniBand

Slurm

Kubernetes

Job description

Fuse Energy, LLC seeks an experienced CUDA performance engineer to design, optimize, and deploy high-throughput CUDA kernels across multi-GPU systems. You will profile hardware bottlenecks, develop tooling to monitor energy usage, and collaborate with ML engineers to integrate kernels into scalable training and inference pipelines.

The role requires deep GPU architecture knowledge, strong C++/CUDA skills, and experience with Nsight, NCCL, and MPI.