20 hours ago
Zürich, SwitzerlandSenior
Responsibilities
- Profile, analyze, and optimize GPU-accelerated applications, including CUDA kernels, memory movement, concurrency, and end-to-end throughput.
- Improve performance through CUDA C++ kernel tuning, GPU libraries, multi-GPU and multi-node scaling, MPI/NCCL communication, and CPU/GPU overlap.
- Build reproducible benchmarks, performance reports, tuning recommendations, reference implementations, examples, patches, and reusable internal libraries.
- Debug correctness and performance issues and advise customers on deployment best practices involving containers, schedulers, and clusters.
- Collaborate with customer, partner, product, and engineering teams to validate fixes and influence the NVIDIA roadmap for energy workloads.
Requirements
- BS/MS or equivalent experience in computer science, computer engineering, electrical engineering, physics, applied mathematics, or a related field.
- Strong programming skills in C/C++ and Python on Linux.
- Hands-on CUDA programming and GPU performance optimization experience.
- Experience using profiling and debugging tools such as NVIDIA Nsight Systems and Nsight Compute.
- Understanding of parallel computing and performance fundamentals, including vectorization, threading, NUMA, and memory bandwidth/latency.
- At least 5 years of relevant experience in GPU/HPC optimization with a record of delivering speedups and scaling improvements.
- Preferred experience with MPI, distributed systems, multi-node performance tuning, energy/HPC workloads, CI and performance regression testing, Docker, Apptainer, Slurm, and AI workflows alongside simulation.
- Ability to communicate technical findings clearly to engineers and non-engineers and lead performance reviews with customer stakeholders.
About Nvidia
Nvidia designs and sells GPUs and accelerated computing platforms for data centers, AI/ML, graphics, gaming, and automotive, monetizing through hardware, software platforms (CUDA, AI frameworks), and systems like DGX and networking. Customers include cloud providers, enterprises, researchers, and OEMs. Founded in 1993 and headquartered in Santa Clara, it is a public company traded on NASDAQ under NVDA.