Deploy and manage orchestration tools such as NVIDIA Base Command Manager, Slurm, Kubernetes, and related cluster management platforms. Execute benchmarking, validation, and performance optimization activities utilizing tools such as NCCL, HPL, MLPerf, IO500, and related technologies.
Provide expert consulting on AI and HPC infrastructure strategy, planning, implementation, and operational readiness. Collaborate with clients to assess technical requirements and deliver customized infrastructure solutions.
Troubleshoot and resolve complex performance bottlenecks across compute, storage, and network layers. Collaborate with clients to assess technical requirements and deliver customized infrastructure solutions.
Develop comprehensive documentation including architecture diagrams, deployment guides, technical designs, and operational procedures. Act as a trusted technical advisor throughout the customer engagement lifecycle.
Essential Requirements
Proven success in designing and deploying large-scale HPC and AI clusters utilizing NVIDIA, AMD, and Intel technologies.
Strong experience with Linux systems administration, troubleshooting, and automation. Extensive hands-on experience with NVIDIA Base Command Manager, Slurm, Kubernetes, or equivalent orchestration platforms.
Experience with high-performance networking technologies including InfiniBand and RoCE. Experience performing system benchmarking, performance validation, and optimization activities.
Demonstrated ability to manage multiple projects independently and deliver results within dynamic environments. Strong customer-facing communication and consulting skills.
Desirable Requirements
Experience with NVIDIA DGX and DGX SuperPOD platforms. Familiarity with AI/ML and Generative AI infrastructure environments. Knowledge of containerized AI platforms and software stacks.
Experience with Python, Ansible, Terraform, or similar automation technologies. Knowledge of data center operations, networking, storage architectures, and cloud integration methods. Experience supporting GPU clusters comprising thousands of accelerators.