gpu-operator

Production-grade diagnostic automation, preflight validation, Xid error triage, and reproducible test harnesses for NVIDIA AI Enterprise (NVAIE) on Kubernetes & Hybrid Cloud.

Production engineering handbook and reference Kubernetes manifests for AI inference at scale: GPU Operator, MIG/MPS partitioning, Volcano, vLLM, KServe, and DCGM.

Open-source reference platform for learning and deploying Kubernetes & OpenShift AI Infrastructure, and NVIDIA GPU Operator

Shared CI/CD workflows, pytest test framework, and utilities for ROCm GPU Operator components

GPU-aware Kubernetes infrastructure reference with Terraform, NVIDIA GPU Operator, scheduling controls and operational validation.