Directory / NVIDIA

NVIDIA-Certified Professional: Agentic AI

One of the few professional exams dedicated to agents. Only a small share is NVIDIA-specific, which makes it unusually portable for a vendor certification.

What it signals

Practical assessment: Scenario-based multiple choice. Remotely proctored multiple-choice.
Identity assurance: Proctored with identity verification. Online, remotely proctored (Certiverse)
Depth: Professional. Professional level.
Experience expected: About 1 year. 1-2 years recommended.
Portability: Mostly transferable. Only 7% of the blueprint is NVIDIA-specific.
Verifiability: Public badge or registry. Credly badge and public directory.

For hiring managers

One of the few agent-specific professional exams and only 7% is NVIDIA-platform specific, so it is unusually portable for a vendor certification. Still multiple-choice only; pass/fail with no score.

Verifying it

Credly digital badge (accepted by the candidate) plus NVIDIA's public certification directory on Credly: https://www.credly.com/organizations/nvidia/directory

What it covers

Published exam outline

Agent Architecture and Design 15%
Agent Development 15%
Evaluation and Tuning 13%
Deployment and Scaling 13%
Cognition, Planning, and Memory 10%
Knowledge Integration and Data Handling 10%
NVIDIA Platform Implementation 7%
Run, Monitor, and Maintain 5%
Safety, Ethics, and Compliance 5%
Human-AI Interaction and Oversight 5%

By topic

GenAI apps: Some (10 to 20%). Knowledge Integration (RAG, prompts) 10% plus parts of Evaluation and Tuning.
Agents: Core focus (over 35%). Entire exam is about agents; architecture, development, cognition/memory alone total 40%.
Classical ML: Not covered. No model training content.
Operations: Substantial (20 to 35%). Deployment and Scaling 13% + Run/Monitor/Maintain 5% + NVIDIA platform 7% = 25%.
Data: Touched on (under 10%). Data handling is part of a 10% section.
Responsible AI: Some (10 to 20%). Safety/Ethics 5% plus Human-AI oversight 5%.

Exam details

Who the provider says it is for. Software developers, software engineers, solutions architects, ML engineers, data scientists, AI strategists, AI specialists

Format: Multiple choice (count given as a range by NVIDIA)
Delivery: Online, remotely proctored (Certiverse)
Prerequisites: None formal
Recommended experience: 1-2 years in AI/ML roles with hands-on work on production-level agentic AI projects; strong knowledge of agent development, architecture, orchestration, multi-agent frameworks, evaluation, observability, deployment and guardrails
Renewal: Retake the exam

Preparing

Not confirmed

These details were not available from an official source at the time of research.

passingScore (not published)
Whether exam items include code snippets (not stated officially)
Original launch date

Sources (2)