Free Quizzes

Free AWS Certified Machine Learning Engineer - Associate Practice Questions

20 exam-style AWS Certified Machine Learning Engineer - Associate questions with an explanation after every answer. No timer, no signup — about 15 minutes.

About the AWS Certified Machine Learning Engineer - Associate exam

The AWS Certified Machine Learning Engineer - Associate (AIF-C01) is a Foundational-level certification. It covers AI/ML fundamentals, generative AI, foundation models, prompt engineering, responsible AI, and security governance on AWS. The exam has 85 questions and runs 120 minutes. You need a scaled score of 700 out of 1000 to pass.

The 20 practice questions on this page are written by a 12x AWS certified engineer and balanced across the exam domains using the official exam guide weighting. They use the same scenario-based style as the real AIF-C01 exam.

Related AWS Certified Machine Learning Engineer - Associate Practice Questions

File Formats & Parquet — MLA-C01 Practice Question
Data Preparation for Machine Learning
AI Services & Custom Model — MLA-C01 Practice Question
ML Model Development
Endpoint Selection & Real Time Inference — MLA-C01 Practice Question
Deployment and Orchestration of ML Workflows
Drift & Model Monitor — MLA-C01 Practice Question
ML Solution Monitoring, Maintenance, and Security
Kinesis & Streaming Ingestion — MLA-C01 Practice Question
Data Preparation for Machine Learning
SageMaker Built IN Algorithms & Xgboost — MLA-C01 Practice Question
ML Model Development

Sample AWS Machine Learning Engineer Associate questions

Three examples from the 20-question quiz. Answers stay hidden until you reveal them.

Sample Question 1ML Model Development

A startup building a customer sentiment analysis feature for automotive dealership reviews wants to use a pre-trained large language model without writing custom training code, and wants to browse and deploy the model directly from the SageMaker Studio UI. Which AWS service best fits this need?

ASageMaker JumpStart
BSageMaker Ground Truth
CSageMaker Feature Store
DSageMaker Debugger
Show answer
Answer: A — SageMaker JumpStart

SageMaker JumpStart provides a UI-based catalog of pre-trained foundation and open-source models that can be deployed or fine-tuned with a few clicks, exactly matching the no-custom-code requirement.

Sample Question 2Deployment and Orchestration of ML Workflows

A food-delivery platform's restaurant-recommendation model is invoked from serverless inference during lunch and dinner rush hours (11am-1pm and 6pm-9pm), reaching 3,000 requests per minute, then dropping to near zero overnight. The team deployed SageMaker Serverless Inference to save cost, but during the first minutes of each rush period they are seeing elevated p99 latency and some timeouts. What is the MOST likely cause, and what should they do?

AServerless Inference does not support auto scaling, so they must switch to Batch Transform for rush periods
BCold starts occur when serverless concurrency scales up from near-zero traffic; they should switch to a real-time endpoint with Application Auto Scaling and scheduled scaling actions ahead of known rush windows
CThe model artifact is too large for Serverless Inference and must be compressed
DThey should enable Multi-Model Endpoints to reduce the number of containers loaded at rush start
Show answer
Answer: B — Cold starts occur when serverless concurrency scales up from near-zero traffic; they should switch to a real-time endpoint with Application Auto Scaling and scheduled scaling actions ahead of known rush windows

Serverless Inference scales containers on demand, and a sudden jump from near-zero to thousands of requests per minute triggers cold starts (container provisioning + model load) that spike latency. Because the rush windows are predictable, a real-time endpoint with scheduled scaling actions (Application Auto Scaling) can pre-warm capacity ahead of known traffic patterns, eliminating cold-start latency.

Sample Question 3ML Solution Monitoring, Maintenance, and Security

A social media platform's content-ranking model was evaluated at launch and showed no bias across age groups. Six months later, the compliance team wants automated, ongoing detection of any emerging disparity in prediction outcomes across the age-group facet, without manually re-running batch bias analysis each month. Which SageMaker capability should they configure?

AA Model Monitor Data Quality monitor scheduled hourly
BA Model Monitor Bias Drift monitor, which uses SageMaker Clarify to continuously check bias metrics against a baseline
CA Model Monitor Feature Attribution Drift monitor only
DA one-time SageMaker Clarify processing job re-run manually each quarter
Show answer
Answer: B — A Model Monitor Bias Drift monitor, which uses SageMaker Clarify to continuously check bias metrics against a baseline

SageMaker Model Monitor's Bias Drift monitor type integrates with Clarify to continuously compute bias metrics (like Difference in Positive Proportions) on live traffic and compares them to a bias baseline, alerting on drift — exactly the ongoing, automated capability described.

Want more practice?

Sign up free to access our full 500+ AWS Certified Machine Learning Engineer - Associate question bank with quiz history tracking.