(Summary generated by AI based on the full job description)
The project focuses on high‑performance scientific and engineering GPU computing. Required stack includes CUDA, C++ and Python, plus distributed communication: MPI, NCCL, NVSHMEM. Core responsibilities cover designing and optimizing CUDA kernels, performance analysis and bottleneck resolution using profiling tools, implementing multi‑GPU and distributed solutions, developing high‑performance C++/Python code, and debugging memory and performance issues. Experience in seismic/RTM/FWI and the Oil & Gas domain is a strong plus.

Optional


By clicking "Aplikuj" you confirm that you've read and accepted our Terms and Conditions.
This is how the employer processes your data
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Need more information?