Nvidia Senior Systems Engineer – High-Performance AI Networking Applications 

United States, Texas 

333984182

What you will be doing:

• Collaborate with networking teams to plan, implement, and evaluate performance benchmarks on NVLINK, NVSwitch, and InfiniBand powered infrastructures.

• Assess findings and work closely with framework, hardware, and support teams to improve system performance across various deep learning workloads.

• Act as a primary resource for fixing networking and hardware integration issues, focusing on scalable multi-node systems.

• Maintain high communication standards across multiple engineering, support, and R&D teams, ensuring technical and performance goals are met.

• Offer technical mentorship and documentation for internal teams and external partners on standard methodologies in HPC networking deployments.

• Share insights on improving networking strategies for substantial AI and deep learning infrastructure.

What we need to see:

• BS/MS or PhD in Computer Science, Engineering, or related field, or equivalent experience.

• 8+ years of proven experience in AI/HPC Infrastructure.

• Familiarity with AI/HPC job schedulers and orchestrators like Slurm, K8s, or LSF. Practical exposure to AI/HPC workflows employing MPI and NCCL.

• Familiarity with High-Speed Networking pertaining to HPC including InfiniBand, RDMA, RoCE, and Amazon EFA.

• Essential to have an understanding of PyTorch, MegatronLM, and Deep Learning Inference frameworks such as vllm/sglang.

• Proven experience with InfiniBand, NVLINK, and high-speed networking technologies in HPC or large-scale datacenter environments.

• Investigating and evaluating performance in multi-node systems, especially in deep learning or scientific computing tasks.

• Strong analytical, debugging, and technical communication skills.

• Comfortable working in collaborative, multi-faceted teams.

Ways to stand out from the crowd:

• Mastery in deep learning frameworks or distributed training systems.

• Familiarity with datacenter automation, advanced network protocols, and supporting large HPC or AI clusters in production environments.

• Understanding of fast, distributed storage systems like Lustre and GPFS for AI/HPC workload.

• Experience with networking and communications libraries like NCCL, NIXL, NVSHMEM, UCX.

• Experience developing or maintaining cluster management and monitoring tools Ex: ansible for infrastructure as a service, prometheus and grafana for monitoring.