A four-week study plan for the AWS AI Practitioner exam

The AWS Certified AI Practitioner (AIF-C01) exam is foundational: no coding, no maths, and AWS describes the ideal candidate as someone with up to six months of exposure to AI and ML on AWS. That makes it a good fit for a short, focused study plan.

Here's a four-week plan built around the exam's own weightings. If you have more time, stretch each week. If you have less, keep the order and shorten each step.

Start by finding your gaps

Before week one, take CertBrain's free diagnostic. It asks questions across all five domains and tells you which one is weakest. That's where any spare time in the plan should go.

How the plan is weighted

The exam guide splits scored questions across five domains:

Domain	Share of scored questions
1. Fundamentals of AI and ML	20%
2. Fundamentals of Generative AI	24%
3. Applications of Foundation Models	28%
4. Guidelines for Responsible AI	14%
5. Security, Compliance and Governance for AI Solutions	14%

Domains 2 and 3 together are more than half the exam, so the plan gives them the most time.

The daily habit: ten to fifteen minutes of review

Whatever week you're in, start each study session by clearing the cards and questions due for review. CertBrain schedules these with spaced repetition: each one comes back after a longer gap every time you get it right, and sooner when you don't. That short daily review is what stops week one's material fading by week four.

Week 1: Domain 1, the fundamentals

• What AI, machine learning, deep learning and generative AI are, and how they nest inside each other. (We covered this in How they fit together.)
• Supervised, unsupervised and reinforcement learning, and how to spot each in a scenario.
• When AI is and isn't the right tool.
• The stages of a machine learning project, from collecting data to monitoring a model in production.

Week 2: Domain 2, generative AI

• Foundation models and large language models: what they are and how they differ from traditional ML.
• Tokens, embeddings and prompts.
• What generative AI is good at, and its limits: hallucinations, non-determinism, cost.
• The AWS services for building with generative AI, starting with Amazon Bedrock.

Week 3: Domain 3, putting foundation models to work

The biggest domain, so it gets a full week.

• Choosing a model for a use case, and the trade-offs between size, cost and speed.
• Prompt engineering techniques: zero-shot, few-shot and chain-of-thought.
• Retrieval-augmented generation and fine-tuning, and when each one fits. (See Prompting, RAG or fine-tuning.)
• How to judge a model's output, with both automated metrics and human evaluation.

Week 4: Domains 4 and 5, then review

• First half: responsible AI. Bias and fairness, transparency and explainability, and the tools AWS offers to detect bias and add safeguards.
• Second half: security and governance. The shared responsibility model applied to AI, controlling access with IAM, protecting data, and governance and compliance.
• Last few days: mixed review. Take the diagnostic again to see what's changed, spend extra time on your lowest domain, and work through mixed scenario questions.

On exam day

A few rules from the exam guide worth remembering:

• There's no penalty for guessing, so never leave a question blank.
• You pass on your overall score. You don't need to pass each domain separately.
• 15 of the 65 questions are unscored and you can't tell which, so treat every question as if it counts.

For the full rundown of cost, timing and question types, see What is the AWS AI Practitioner exam?. Always check AWS's exam guide before you book, as AWS updates it from time to time.

Ready to start week one? Take the free diagnostic and see where you stand.

CertBrain is an independent study resource. It isn't affiliated with or endorsed by Amazon Web Services. AWS and AWS Certified AI Practitioner are trademarks of Amazon.com, Inc. or its affiliates.