Nine exams from the Linux Foundation, NVIDIA, AWS, Google Cloud, Microsoft and Databricks, grouped by what you want to do: run AI infrastructure, engineer ML systems, or build with GenAI and agents.
Linux Foundation · Associate
The base layer: most production AI workloads run on Kubernetes, and this exam is performance-based rather than multiple choice.
For: DevOps and platform engineers
Skills: Kubernetes administration, container orchestration
NVIDIA · Associate
Covers GPU infrastructure and AI operations fundamentals in the NVIDIA ecosystem.
For: MLOps and AI infrastructure staff
Skills: GPU infrastructure, NVIDIA stack, AI operations
Heads-up: Pass mark 70%.
Official exam page →AWS · Foundational
Accessible entry point to AI/ML and generative AI concepts, with a responsible-AI component.
For: IT professionals, analysts, product managers
Skills: AI/ML basics, foundation models, prompt engineering, Bedrock and SageMaker
Heads-up: Free practice exams on AWS Skill Builder.
Official exam page →AWS · Associate
Building, deploying and monitoring ML solutions on AWS. About a year of SageMaker or AWS ML experience is expected.
For: Backend, DevOps, data and MLOps engineers
Skills: Data prep, model development, deployment and orchestration, monitoring and security
Heads-up: Changing: AWS says the last day to take MLA-C01 in English was 28 Sep 2026; the updated MLA-C02 is in beta. Check AWS before booking.
Official exam page →Google Cloud · Professional
Google's most demanding ML credential, blending classical ML with generative AI on Google Cloud. Recommended: 3+ years of industry experience.
For: Experienced ML engineers
Skills: BigQuery ML, AutoML, pipelines, serving, monitoring
Microsoft · Associate
Operationalising ML models (MLOps) and generative AI applications (GenAIOps) on Azure.
For: Data scientists, DevOps and ML engineers
Skills: MLOps, GenAIOps, automation and monitoring
Heads-up: Renewed through a free online assessment.
Official exam page →NVIDIA · Associate
Transformers, LLM deployment and the NVIDIA, LangChain and Hugging Face tooling around them.
For: AI DevOps, cloud and ML engineers
Skills: Transformer architecture, LLM deployment, Triton, LangChain
Heads-up: NVIDIA's own course materials for this exam are reported as retired.
Official exam page →Databricks · Associate
Building RAG applications and LLM workflows on Databricks. About six months of hands-on GenAI experience is expected.
For: ML engineers building LLM apps
Skills: RAG design, Model Serving, MLflow, Unity Catalog, Vector Search
Heads-up: Python and SQL.
Official exam page →Linux Foundation · Associate
One of the first certifications for MCP, the standard for connecting AI agents to tools and data.
For: Anyone building agents and tool integrations
Skills: MCP clients, servers, manifests and capabilities
Heads-up: Now available; the source article listed it as launching soon with no price. Price and length taken from the Linux Foundation page. One retake included.
Official exam page →Fees as reported in August 2026; confirm on the official page.
| Certification | Provider | Level | Cost | Length |
|---|---|---|---|---|
| CKA — Certified Kubernetes Administrator | Linux Foundation | Associate | $445 | 2 hours |
| NVIDIA-Certified Associate: AI Infrastructure & Operations | NVIDIA | Associate | $125 | 60 min |
| AWS Certified AI Practitioner (AIF-C01) | AWS | Foundational | $100 | 90 min |
| AWS Certified Machine Learning Engineer – Associate | AWS | Associate | $150 | 130 min |
| Google Cloud Professional Machine Learning Engineer | Google Cloud | Professional | $200 | 2 hours |
| Microsoft Certified: Machine Learning Operations Engineer Associate | Microsoft | Associate | $165 | 120 min |
| NVIDIA-Certified Associate: Generative AI LLMs | NVIDIA | Associate | $125 | 1 hour |
| Databricks Certified Generative AI Engineer Associate | Databricks | Associate | $200 | 90 min |
| Model Context Protocol Associate (MCPA) | Linux Foundation | Associate | $250 | 90 min |
Choosing the right exam.
If you are new to AI, the AWS Certified AI Practitioner ($100) is the gentlest entry. If you already run infrastructure, the Certified Kubernetes Administrator is the base that most production AI workloads sit on.
The source article is written for them. CKA, NVIDIA's AI Infrastructure and Operations associate, AWS Machine Learning Engineer, and Microsoft's AI-300 MLOps exam map most closely to infrastructure and operations work.
No. The source article's advice is to pick the one that fits your career goal instead of collecting credentials. Practical experience matters more than the number of badges.
Prices come from the source article, dated August 2026, except where noted. Exam fees and formats change, so confirm on the official page before booking.
The Model Context Protocol Associate from the Linux Foundation tests understanding of MCP, the standard for connecting AI agents to external tools and data. It is now listed at $250 for a 90-minute exam.
AWS says MLA-C01 was retired for English on 28 September 2026 and the updated MLA-C02 is in beta. Read the AWS page for the current state before you register.
