CUDA Engineering Expert - Remote

We are looking for a seasoned CUDA Engineering Expert to join a remote GPU‑optimization effort. The contractor will apply deep knowledge of CUDA, C++ and GPU programming to make high‑performance kernels faster and more maintainable. Day‑to‑day work involves profiling and tuning GPU kernels, spotting performance bottlenecks and crafting targeted improvements. You will refactor C++ and CUDA code, write shader logic in GLSL and WebGPU, and record the optimisation steps and results. Collaboration with a distributed, cross‑functional team and participation in architecture discussions are also part of the role. The position requires proven CUDA and GPU kernel optimisation skills, strong C++ expertise, practical GLSL and WebGPU experience, and familiarity with profiling tools such as NVIDIA Nsight. A solid grasp of GPU performance characteristics and architecture, excellent analytical and communication abilities, and the capacity to thrive in a remote setting are essential. Experience with high‑performance computing, multi‑GPU architectures, or graphics/AI acceleration is considered a plus. What the role asks for: - Strong CUDA programming and GPU kernel optimization expertise - Advanced C++ development skills - Hands‑on experience with GLSL and WebGPU - Experience with GPU profiling tools such as NVIDIA Nsight - Deep understanding of GPU performance and architecture - Excellent analytical, problem‑solving and technical communication abilities - Ability to work effectively in a remote setting - Nice‑to‑have: background in HPC, multi‑GPU architecture, graphics or AI acceleration