AWS Certified AI Practitioner (AIF-C01)

Learning plan

About 30 hours of focused study, split into 4 phases. Pick your weekly pace and the timeline adjusts.

Estimated finish: 3 weeks

What the exam covers

Source: official exam guide. Weights can change, so check before you start.

Before you start

• A free AWS account (Free Tier) so you can click through Bedrock and SageMaker consoles
• Basic cloud vocabulary (regions, IAM, S3). AWS Cloud Practitioner knowledge helps but is not required

1. Week 1 · 8 h

   AI/ML and GenAI fundamentals

   • Explain supervised vs unsupervised vs reinforcement learning and when each fits
   • Know the ML lifecycle (data prep, training, evaluation, deployment, monitoring) and core metrics (accuracy, precision/recall, F1, AUC)
   • Define tokens, embeddings, context windows, transformers, foundation models and hallucination

   Use: AWS Skill Builder – AIF-C01 Exam Prep Plan (free) · AIF-C01 official exam guide

2. Weeks 1–2 · 10 h

   Foundation models on AWS (biggest domain)

   • Choose between prompt engineering, RAG, fine-tuning and continued pre-training, including cost/effort trade-offs
   • Know Amazon Bedrock (model choice, Knowledge Bases, Agents, Guardrails) and when to use SageMaker AI / JumpStart instead
   • Recognise the managed AI services (Comprehend, Rekognition, Transcribe, Translate, Textract, Lex, Kendra, Q) and match them to use cases
   • Understand inference parameters (temperature, top-p, max tokens) and evaluation methods (ROUGE, BLEU, BERTScore, human eval)

   Hands-on: In the Bedrock console, build a Knowledge Base over a few PDFs, query it, then add a Guardrail that blocks a topic. Compare answers at temperature 0 vs 1.

   Use: Amazon Bedrock User Guide

3. Weeks 2–3 · 6 h

   Responsible AI, security and governance

   • Name the responsible-AI dimensions (fairness, explainability, privacy, robustness, transparency) and the AWS tools for each (SageMaker Clarify, Model Monitor, A2I, Guardrails)
   • Apply the shared responsibility model, IAM, encryption, PrivateLink and data-residency basics to AI workloads
   • Know governance artefacts such as model cards, data lineage and audit logging with CloudTrail

   Use: AWS Responsible AI resources

4. Week 3 · 6 h

   Practice exams and gap review

   • Take the free official practice question set, then one full-length timed practice exam
   • Review every miss and map it to a domain; re-study the two weakest domains
   • Retake until you consistently score 80% or higher

   Use: AWS Official Practice Question Set (Skill Builder) · Tutorials Dojo AIF-C01 practice exams

You're ready to book when…

• You can explain when to use RAG vs fine-tuning vs prompt engineering without notes
• You can match each AWS AI service to a scenario in under 10 seconds
• You score 80% or higher on two different full-length practice exams

Exam tips

• Questions are scenario-based; look for the most managed, least-effort option that meets the requirement.
• Unanswered questions count as wrong, so answer everything and flag the ones to revisit.