GPU / CUDA Developer – Long-term Freelance Project

Jan 14, 2025

widget pic
widget pic

We’re looking for an experienced GPU/CUDA developer to join a major long-term project. This is a rare chance to work on performance-critical systems, building and optimizing GPU-accelerated solutions in collaboration with a leading engineering team.

Content

Allocation: 100 %
Start: Year-end (estimate)
Duration: Long-term, major case for the client’s client
Locations: Tampere, Turku, Helsinki, Tallinn, Szczecin, Wroclaw (other EU locations possible)
Topic: GPU parallel computing / CUDA + C++

Must-have skills

  • Strong C++ expertise

  • Hands-on CUDA development (kernels, memory management, parallelization)

  • Profiling and performance optimization (e.g. Nsight, Visual Profiler)

  • Experience integrating GPU-powered modules into larger systems

  • Solid testing and debugging practices

Nice-to-have

  • OpenCL, Vulkan Compute, Metal Compute

  • HPC / parallel computing (MPI, OpenMP)

  • TensorRT, PyTorch/TensorFlow GPU optimization

  • Background in performance-critical software

This role is not super urgent — the project is expected to kick off around the end of the year — but we are already mapping interested candidates so the team is ready when it starts.

Rate: Offer | Nordkood adds +6 % on top