AWS Certified AI Practitioner
AWS's entry exam for AI and generative AI, aimed at people who use AI tools on AWS rather than build models. Practice what it tests: core ML and GenAI concepts, picking the right AWS service for a business problem, RAG and prompt engineering with Amazon Bedrock, responsible AI, and securing and governing AI systems. Every question is original and maps to AWS's current exam guide.
What you'll study
Every section follows the official AIF-C01 exam outline. Click a domain to see its sections.
Basic AI terms and how AI, ML, deep learning, GenAI and agentic AI relate, data and learning types, inferencing, when AI fits and when it doesn't, AWS managed AI services, and the ML pipeline, MLOps and model metrics.
Tokens, embeddings and model types, GenAI use cases, the foundation model lifecycle, context engineering, agentic AI and MCP, GenAI strengths and weaknesses, choosing models, business value, AWS GenAI services and token pricing tradeoffs.
Choosing a foundation model and its inference parameters, RAG and vector databases, customization approaches and AI agents, prompt techniques, best practices and risks, fine-tuning and data preparation, and evaluating models and FM apps against business goals.
Features of responsible AI, Amazon Bedrock Guardrails, responsible model selection, legal risks of GenAI, dataset characteristics, bias, variance and overfitting, transparent versus opaque models, model cards and explainability tools, and human-centered design.
AWS services that secure AI systems, source citation and data lineage, secure data engineering, AI security and privacy, hallucination detection and grounding, and the governance and compliance services, strategies and frameworks for AI.
Study, Quiz, Exam
Read the concepts, practice in short rounds, then test yourself on a mock exam with targeted feedback on where you need to focus more. Every question is original and maps to the official AIF-C01 outline.
Learn the concept first
Short study cards cover each section before you answer a single question.
An algorithm is a learning procedure. Training runs it on example data, and the output is a trained model, the artifact that actually makes predictions. Inferencing is the later step of sending new data to that deployed model and getting a prediction back.
| Term | What it is |
|---|---|
| Algorithm | The procedure, such as XGBoost; same whatever data it sees |
| Training data | The examples the algorithm learns from |
| Features | The input columns fed to the model |
Ten-question rounds
Every answer explained. Score 80% and the next section unlocks.
A bank trained a fraud detection model last quarter and deployed it. Now, each time a customer pays by card, the transaction details are sent to the model, and the model returns a fraud score.
What is this use of the deployed model called?
Inferencing means using a trained model to make predictions on new data, such as scoring each new card transaction. Training is the earlier step in which an algorithm learns from example data to produce the model. Hyperparameter tuning searches for the best training settings, and feature engineering prepares input data, and both happen before the model goes into use.
Know exactly where to focus
Timed practice exams any time, with a study plan built from your mistakes.
What to expect on the AIF-C01
Made for anyone who works with AI on AWS or wants to, including people who use AI tools without building them. You don't need to be a developer or a data scientist.
Source: AWS Certified AI Practitioner, checked October 16, 2026
Where this cert takes you
Real job titles where AI Practitioner skills are the day-to-day work.
- Business Analyst
- Junior Systems Analyst
- Data Analyst
- Cloud Support Associate
- IT Business Analyst
- Technical Support Engineer
- Systems Analyst
- AI Business Analyst
- Solutions Engineer
- Cloud Engineer
- Implementation Consultant
- Technical Product Owner
- Senior Systems Analyst
- AI Solutions Consultant
- Solutions Architect
- AI Product Manager
- AI Governance Lead
- Technical Program Manager
Ranges based on BLS Occupational Employment & Wage Statistics (May 2025) for Computer Systems Analysts: 10th percentile $67,340, median $105,850, 90th percentile $167,710. Adjusted by metro. AI Practitioner tests choosing and applying AI services for business needs, which is the work of computer systems analysts; specialist ML engineer and data scientist roles are counted under other occupations. The first role in each column is the most common entry point.
Recommended ProCourse
A good course to pair with your cert study for learning hard tech skills while you study for your cert
AI Engineering & Automation
Go beyond prompting. Call APIs, build tools that classify and generate, wire up RAG pipelines, create autonomous agents with memory and tool use, then train and deploy your own models. AI engineering, not AI tourism.
View CurriculumQuestions
Start practicing for AIF-C01
Included with an opnf.sh subscription, $15/month or $120/year.
Get CertPrepAmazon Web Services, AWS, AWS Certified AI Practitioner, Amazon Bedrock and Amazon SageMaker AI are trademarks of Amazon.com, Inc. or its affiliates. opnf.sh is not affiliated with, sponsored by or endorsed by Amazon Web Services. All practice questions are original and are not taken from any Amazon Web Services exam.