Learning path·Intermediate · 35–50 hours

Become an AI Infrastructure Engineer with real Kubernetes and real GPUs

Run GPU clusters that don't melt down. Kubernetes, GPU Operator, MIG, MPS, observability.

0 interactive lessons16 GPU labs
6
Modules
16
Hands-on GPU labs
2
Cert checkpoints

About this path

AI Infrastructure Engineers run the GPU clusters that everyone else's models depend on. The role lives at the intersection of Kubernetes platform engineering and GPU-specific operations, and almost no online course teaches it through real clusters. This path does. Every module is a lab on a live, isolated Kubernetes environment with real GPU scheduling, real GPU operator components, and real triage scenarios that break the way they break in production.

Skills you'll put on a resume

  • Schedule GPU workloads on Kubernetes with the right resource requests, limits, and priority classes
  • Install and triage the NVIDIA GPU Operator end-to-end, attributing every component's role
  • Share single GPUs across workloads with CUDA streams, MPS, and MIG, and know when each fits
  • Implement the four-stage cost-audit pipeline (measure → classify → price → recommend) for GPU fleets
  • Profile PyTorch training with Nsight Systems and the built-in profiler to find real bottlenecks
  • Pass the NVIDIA-Certified Associate AI Infrastructure & Operations exam

For

Platform engineers, SREs, and DevOps engineers responsible for running GPU workloads in production. Comfortable with Linux and Kubernetes basics; new to GPU-specific operations.

Prerequisites

  • Comfortable with Linux command line and basic shell scripting
  • Kubernetes basics (Pods, Deployments, Services), equivalent of CKA prep
  • Familiarity with containers (Docker / containerd)

Guides & articles

Deep-dive reading that pairs with this course

NCA-AIIO Complete Guide 2026 — NVIDIA AI Infrastructure & Operations

Master the NVIDIA Certified Associate - AI Infrastructure and Operations (NCA-AIIO) certification. Exam domains, study plan, career impact, and preparation strategies for data center and IT infrastructure professionals.

Read

NCA-AIIO Exam Domains: Complete Breakdown of All 3 Domains (2026)

Every NCA-AIIO exam domain explained topic by topic: Essential AI Knowledge (38%), AI Infrastructure (40%), and AI Operations (22%), with the DGX, NVLink, InfiniBand, DCGM, and MIG concepts NVIDIA tests.

Read

How to Pass NCA-AIIO on Your First Attempt (2026 Strategy)

A first-attempt strategy for the NVIDIA NCA-AIIO AI Infrastructure & Operations exam: where candidates lose points, how to prep for a fast 50-question associate exam, and the final-week protocol.

Read

NCA-AIIO Study Plan: 4-Week Preparation Schedule (2026)

A complete 4-week study schedule for the NVIDIA NCA-AIIO AI Infrastructure & Operations exam: weekly goals across all three domains, hands-on GPU labs, and a practice-exam cadence that peaks on exam day.

Read

NVIDIA NCA-AIIO Cheat Sheet 2026: Key Concepts & Decision Rules

Fast-review reference for the NVIDIA NCA-AIIO exam: the software stack, NVLink vs InfiniBand, DGX vs HGX, training vs inference, DCGM and MIG, and the facts a 50-question associate exam tests.

Read

NCP-AII Complete Guide 2026: NVIDIA AI Infrastructure Certification

Master the NVIDIA Certified Professional - AI Infrastructure (NCP-AII) certification. Exam domains, DGX bring-up and cluster validation skills, salary data, and a preparation path for data center engineers building GPU clusters.

Read

How to Pass NCP-AII on Your First Attempt (2026 Strategy)

A first-attempt strategy for the NVIDIA NCP-AII AI Infrastructure exam: where candidates lose points across bring-up and validation, how to prepare without DGX hardware, and how to run the final week.

Read

NCP-AII Exam Domains: Complete Breakdown of All 5 Domains (2026)

Every NCP-AII exam domain explained topic by topic: cluster test and verification with HPL and NCCL, DGX bring-up sequencing, Base Command Manager control plane, Xid troubleshooting, and MIG/DPU physical layer management.

Read

NCP-AII Study Plan: 6-Week Preparation Schedule (2026)

A complete 6-week study schedule for the NVIDIA NCP-AII AI Infrastructure exam: weekly goals across bring-up, validation, and control plane, hands-on GPU lab work, and a practice-exam cadence that peaks on exam day.

Read

NVIDIA NCP-AII Cheat Sheet 2026: Key Concepts & Commands

Fast-review reference for the NVIDIA NCP-AII exam: deployment sequence, validation toolchain, Xid error table, DCGM diagnostic levels, MIG vs vGPU comparison, and the BCM control-plane map.

Read

NCP-AIO Complete Guide 2026: NVIDIA AI Operations Certification

Master the NVIDIA Certified Professional - AI Operations (NCP-AIO/NCP-AIOL) certification. New lab-based exam format, all four domains, Slurm and Kubernetes operations skills, salary data, and a preparation path for GPU cluster operators.

Read

How to Pass NCP-AIO on Your First Attempt (2026 Strategy)

A first-attempt strategy for the lab-based NVIDIA NCP-AIO exam: time budgeting across 30 questions and 3 live labs, the terminal fluency bar, where candidates lose points, and the final-week protocol.

Read

NCP-AIO Exam Domains: Complete Breakdown of All 4 Domains (2026)

Every NCP-AIO exam domain explained topic by topic: BCM and Mission Control installation, Slurm and Run:ai administration, training and Triton/NIM inference workloads, and the GPU troubleshooting decision tree.

Read

NCP-AIO Study Plan: 6-Week Preparation Schedule (2026)

A complete 6-week study schedule for the lab-based NVIDIA NCP-AIO exam: weekly goals across installation, administration, workloads, and troubleshooting, daily hands-on GPU reps, and a practice cadence that peaks on exam day.

Read

NVIDIA NCP-AIO Cheat Sheet 2026: Key Concepts & Commands

Fast-review reference for the lab-based NVIDIA NCP-AIO exam: Slurm admin command table, Run:ai quota model, Triton serving features, the five-layer troubleshooting tree, Xid/ECC quick table, and MIG geometry rules.

Read

Ready to start?

Pro gives you all 16 labs in this path, every other lab on Preporato, and every practice test. $29.99/mo, cancel anytime.