studying for the AI1-C01 - AWS certified AI practitioner beta exam



Here’s a summary of how I studied for the AWS Certified AI Practitioner exam.

$ My Background


Several years ago, I was employed in a role that required me to know artificial intelligence (AI) and natural language processing (NLP) really well, so I already had working knowledge of AI before taking the test. Prior knowledge is certainly not required, but I believe mine helped me to hit the ground running quickly with studying.

I had also taken the AWS Certified Solutions Architect - Associate and passed earlier in 2024, so I am very familiar with the structure of AWS tests and have some knowledge of other AWS services.

Additionally, in my previous role, I used AWS frequently and am familiar with the hands-on aspect of AWS.

$ What I Used to Study


Because I had prior knowledge and experience, I almost entirely used Stephane Maarek’s Ultimate AWS Certified AI Practioner AI1-C01 course. I also used the practice tests associated with his exam study course. I didn’t use anything else aside from quick Google searches about individual inquiries.

(Note: I also used his course to study for the AWS Certified Solutions Architect Associate exam, alongside many completed AWS Skill Builder exercises. AWS Skill Builder is another great platform to use for studying for those who need a hands-on component to learn.)

$ General Thoughts About the Exam


The exam was harder than I had expected, honestly.

My studying was focused on AWS’s AI-powered services, but I felt like the exam was geared more toward concepts and fundamentals rather than services. For example, it was more important to know if supervised learning was appropriate versus unsupervised learning rather than being able to know what exact AWS service to use. This is a generalization, however, and many questions do demand you know which service does what.

That being said, the exam is not overwhelmingly challenging, and I think I still would’ve passed in the first or second attempt without prior knowledge and experience in AI.

$ Topics I Remember Covered in My Exam


  • Inference parameters (temperature, Top P, Top K)
  • Batch inference
  • Types of throughput
  • Audit management
  • Embeddings
  • Knowledge bases and RAGs
  • Data splits
  • Machine learning model types
  • Detecting and explaining biases
  • Explainability and interpretability
  • Features
  • Feature transformations and feature engineering
  • Overfitting and underfitting models
  • Metrics
  • Custom Entity Recognition
  • Real-time, asynchronous, and batch
  • Prompt Engineering
  • Model Evaluations
  • Supervised learning
  • Unsupervised learning
  • Self-supervised learning
  • Reinforcement learning
  • Privacy and PII in models
  • Securing assets in a model deployment and IAM
  • How to use Amazon S3 with model deployments
  • Other AI-powered AWS services (i.e. Transcribe, Comprehend, Data Wrangler, Ground Truth, etc.)
  • Various Amazon Q services/facilities
  • Various Amazon SageMaker services/facilities





Written: August 30, 2024


GITHUB / EMAIL