Experience
Ten years of low-level C++ on compute and memory bound workloads. The roles below are the GPU ones; each is described by what ran on the card and what was measured.

Senior HPC Engineer
Jugo
·
Contract
March 2026 to September 2026
·
United Kingdom
·
Remote
Built and optimized custom CUDA kernels for live AI video processing on Jugo's immersive meetings platform, which renders real-time 3D environments in Unreal Engine on cloud GPUs.
Developed the GPU compositing stage that injects game-engine frames into NVIDIA DeepStream and overlays them with live video.
Optimized kernels on RTX PRO 6000 Blackwell, cutting P95 per-frame latency from 25.1 ms to 18.3 ms and reducing jitter under high network latency. Nsight-driven profiling of compute and memory bound stages: occupancy, launch configuration and memory coalescing.
Skills
HIP/ROCm, System Performance, C++, CUDA, High Performance Computing (HPC), GPGPU, Performance Tuning
Co-founder & Technical Lead
NitroSpark AI
·
Full-time
June 2024 to April 2026
·
United Kingdom
·
Hybrid
Co-founded NitroSpark AI and served as Technical Lead.
Skills

GPU Systems Engineer
Union One
·
Contract
August 2018 to August 2024
·
United Kingdom
·
Remote
Delivered the GPU-resident render pipeline for Union One, a tactical FPS: particle systems and related effects computed on the GPU rather than the CPU, with Vulkan driving rendering and the surrounding engine systems. Managed systems under frame budget and memory layout constraints, profiled and optimized to target per-frame time budget.
Skills
System Performance, OpenGL Shading Language (GLSL), C++, GPGPU, Performance Tuning
Tools
C++, CUDA, Vulkan, GLSL and Slang, Python. Nsight Compute and Nsight Systems for profiling, PTX and SASS inspection, cuSPARSE and the NPP libraries as baselines, OpenMP references on the CPU side.
