3 days ago
France (city not specified) or Berlin, GermanySenior
Responsibilities
- Collaborate with NVIDIA training framework developers and product teams to help partners adopt new features.
- Deploy, debug, benchmark, and optimize AI training and inference workloads on large-scale NVIDIA GPU platforms.
- Analyze performance, identify bottlenecks, and share actionable insights with customers and internal teams.
- Resolve cluster performance and stability issues across applications, frameworks, runtimes, schedulers, and hardware.
- Guide customers in scaling workloads efficiently and reliably on current-generation NVIDIA GPUs.
- Help customers implement resiliency features in AI training pipelines as part of Europe’s Sovereign AI initiative.
Requirements
- BS, MS, PhD, or equivalent practical experience in Computer Science, Electrical or Computer Engineering, Physics, Mathematics, or a related engineering field.
- 8+ years of experience with accelerated computing technologies at cluster scale, ideally involving NVIDIA platforms.
- Strong programming skills in at least one of C, C++, or Python.
- Practical experience resolving bottlenecks in large-scale training workloads or parallel applications.
- Hands-on experience profiling and debugging large parallel applications.
- Understanding of CPU and GPU architectures, CUDA, parallel filesystems, high-speed interconnects, cluster scheduling, and resource management.
- Knowledge of training pipelines and frameworks, including their internal operations and performance characteristics.
- Experience debugging production training pipelines across thousands of GPUs is preferred.
- Experience with Nsight Systems, Nsight Compute, NCCL, MPI, and low-level communication libraries is preferred.
- Understanding of PyTorch, Megatron-LM, NeMo, vLLM, Dynamo, TensorRT-LLM, RedHat Inference Server, or SGLang is preferred.
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.
