Senior HPC Engineer
Core
Validating, tuning, and owning the production performance of large-scale NVIDIA GPU clusters for AI training and HPC workloads.
Role type
Senior HPC Engineer (Infrastructure)
Builds
High-performance GPU clusters connected by InfiniBand for AI startups and enterprises
Domain
Cloud Infrastructure / High-Performance Computing / AI
Deliverable
production ML models
Required skills
NVIDIA GPU infrastructure operations, InfiniBand subnet management, Slurm workload management, Linux systems administration, cluster benchmarking and tuning, infrastructure automation
Preferred skills
HGX H100/H200/B200 systems, NVIDIA UFM, GPUDirect RDMA, parallel file systems (WEKA, VAST, Lustre), Kubernetes/OpenStack, MPI-based HPC applications
Technologies
NVIDIA CUDA, DCGM, InfiniBand, Slurm, NCCL, HPL, Git, Linux
Responsibilities
Run burn-in, health checks, and benchmarks (NCCL, HPL) to validate cluster performance before handover; Tune clusters for AI training and HPC workloads including NCCL settings and topology-aware scheduling; Build GPU, fabric, and job-level monitoring and lead incident response; Automate provisioning, configuration, and validation with infrastructure-as-code; Collaborate with Deployment Architects, Network Engineers, and vendors to resolve issues.
Seniority
Senior, hands-on IC