DSX NVCRE Senior Engineer
Core
Build and operate tooling to verify AI infrastructure readiness for production, validating GPU clusters and producing auditable evidence for customers and partners.
Role type
Senior IC software engineer (infrastructure/validation)
Builds
Cluster readiness and certification software, open-source Cluster Readiness Engine, benchmark catalogs, and provenance pipelines.
Domain
AI infrastructure, High-Performance Computing (HPC), GPU clusters, Kubernetes, Slurm
Deliverable
production ML models
Required skills
Go, Python, Linux systems, Kubernetes operators/controllers, HPC schedulers (Slurm), GPU cluster fundamentals (NCCL, InfiniBand/RoCE, DCGM), distributed systems benchmarking, technical investigation
Preferred skills
Large-scale distributed training/inference (PyTorch, Megatron, NeMo, vLLM, SGLang, TensorRT-LLM), NVIDIA ecosystem (DGX, GB200/GB300, CUDA, NVLink, NVSwitch), open-source infrastructure contributions, software supply-chain security/attestation, observability stacks (Prometheus, Grafana, OpenTelemetry), GitOps, infrastructure as code
Responsibilities
Design and implement cluster readiness tooling in Go and Python; build benchmark catalogs for communication, compute, storage, training, and inference; develop provenance pipelines for auditable results; integrate new GPU platforms and schedulers; root-cause performance gaps with cross-functional teams; contribute to open-source projects; author design documents and operational guides
Seniority
Senior, hands-on IC
