thinQmaster Virtual testing partner Adaptive practice exams that find your weak spots and drill them. Every answer explained. From $69
Enroll

thinQmaster

Find your exam

Five questions free, before you decide.

Step 1 · choose a vendor

In one of our camps? Your practice exams come with the tuition — they are on your study page already.

NVIDIA · AI & Machine Learning · Professional

NVIDIA-Certified Professional: AI Infrastructure exam guide

Exam NCP-AII

NCP-AII validates the ability to deploy, configure and validate GPU servers and AI clusters: system bring-up, control plane installation, physical layer management, cluster testing and troubleshooting. It targets the engineers building out AI factories, one of the fastest-growing infrastructure roles in 2026.

Questions70-75
Time limit120 minutes
Passing scorePass/fail (not published)
Exam fee$400 USD
Valid for2 years
DeliveryOnline, remotely proctored (Certiverse account required)

Format: Multiple choice; NVIDIA's page also references a hands-on lab component; pass/fail.

Who it's for

Infrastructure engineers with 2-3 years of hands-on experience deploying GPU servers and AI clusters.

What the exam covers

Cluster Test and Verification33%
System and Server Bring-Up31%
Control Plane Installation and Configuration19%
Troubleshooting and Optimization12%
Physical Layer Management5%

Eligibility & prerequisites

Required: None.

Recommended: 2-3 years deploying, configuring and validating GPU-based servers and AI clusters; recommended training is AI Infrastructure and Operations Fundamentals and the AI Infrastructure Professional Workshop.

Keeping it active

Valid 2 years from issuance. Recertify by retaking the current version of the exam before it expires.

Skills & careers

Skills it validates

  • DGX/HGX server bring-up and firmware updates
  • BMC and out-of-band management
  • Installing Base Command Manager and control plane services
  • Cluster validation with NCCL tests and HPL
  • Cabling and transceiver validation
  • Fabric and GPU troubleshooting

Job roles

AI Infrastructure EngineerData Center Deployment EngineerHPC EngineerField Applications EngineerSystems Integration Engineer

Where it fits in your path

Planning a degree too? See how this cert earns college credit in the College Credit Navigator.

Build my study plan

  1. Weeks 1-2: Server bring-up: BMC setup, firmware and BIOS settings, OS and driver install, DGX OS, NVSM and health checks.
  2. Weeks 3-4: Cluster test and verification: NCCL tests, HPL, ClusterKit-style burn-in, interpreting bandwidth and error results.
  3. Week 5: Control plane: Base Command Manager, Slurm or Kubernetes installation, networking for management and storage.
  4. Week 6: Physical layer and troubleshooting: cabling, transceivers, link errors, thermal issues.
  5. Week 7: Practice questions, focusing on the two largest domains (64% combined).

Make it yours: pick your exam date and weekly hours. We'll split your time by domain weight and give you a printable plan.

Exam-day tips

  • Cluster Test and Verification (33%) and Bring-Up (31%) dominate; know expected NCCL bandwidth patterns and how to read failures.
  • Know the order of operations for bringing a new node into a cluster.
  • Expect questions on BMC/Redfish and out-of-band tasks, not just OS-level commands.
  • Scoring is pass/fail with no published cut score, so aim for consistent strength across all five domains.

Key terms for this exam

All 31 terms Study with thinQcards™

Frequently asked questions

How much does the NCP-AII exam cost?

$400 per attempt, booked through Certiverse.

How long is NCP-AII valid?

Two years from issuance. You recertify by retaking the exam.

Is the exam online?

Yes. NVIDIA delivers its certification exams online with remote proctoring through Certiverse.

NCP-AII vs NCP-AIO?

NCP-AII covers building and validating clusters (bring-up and test); NCP-AIO covers running them day to day (administration, scheduling and troubleshooting).

Is there a prerequisite?

No formal prerequisite, though NVIDIA recommends 2-3 years of hands-on GPU server and cluster deployment experience.

Official page: www.nvidia.com. Reviewed November 2026. Vendors change fees and objectives, so confirm with the vendor before you book. Spot something out of date? Tell us.