How is this hardest-questions list ranked?

When live data is available, the list ranks public diagnostic questions by recent misses and attempt volume. While weekly data fills in, it shows a high-signal fallback set from observed public quiz misses.

What should I do after reviewing these questions?

Read the full explanation for each missed concept, retake the free diagnostic, then save or email the score report so your weak areas turn into a focused study plan.

Rolling 7-day practice signal

Hardest AWS Certified Machine Learning Engineer - Associate questions this week

Q: What are the hardest AWS Certified Machine Learning Engineer - Associate questions this week?

The current review set is focused on File Formats & Parquet, AI Services & Custom Model, Endpoint Selection & Real Time Inference. These are the topics to review before retaking the diagnostic.

These are the public diagnostic questions people are missing most often. Use the list as a weekly study loop: review the concept, then retake the diagnostic.

Weekly review loop

Step 1Start with the most-missed topic and read the short answer explanation.
Step 2Open the full breakdown for the concept before moving to the next topic.
Step 3Retake the diagnostic and email the score report if you want the study plan.

Live

answers analyzed

practice sessions

questions to review

Live weekly data is still filling in, so this page is showing the current high-signal review set.

Weekly report

Send the current weak-area list to yourself, then use the links as a study queue before your next diagnostic attempt.

Email this report

#1 most missed

File Formats & Parquet

Data Preparation for Machine Learning

Review

high-signal topic

A retail analytics team stores 6 months of clickstream logs in S3 as gzip-compressed CSV files. Their SageMaker training jobs spend most of their time scanning full files to select only 8 of 40 columns, and storage costs are climbing. They want a format that supports column pruning and predicate pushdown for faster, cheaper reads. Which file format should they migrate to?

Correct answer

B. Convert the files to Apache Parquet with columnar compression

Parquet is a columnar format that lets query and training engines read only the required columns (column pruning) and skip row groups using min/max statistics (predicate pushdown), directly solving the "read 8 of 40 columns" problem while also compressing better than row-oriented CSV.

Read full explanation Retake diagnostic

#2 most missed

AI Services & Custom Model

ML Model Development

Review

high-signal topic

A retail chain wants to automatically extract product names, prices, and expiration dates from photos of shelf labels. The team has no ML engineers and needs a solution within two weeks. Which approach is MOST appropriate?

Correct answer

B. Use Amazon Textract to extract text from the label images, then apply simple regex parsing

Amazon Textract is a managed AI service purpose-built for extracting text (and key-value pairs) from documents and images. Combined with lightweight regex, it meets the two-week deadline with no ML expertise required.

Read full explanation Retake diagnostic

#3 most missed

Endpoint Selection & Real Time Inference

Deployment and Orchestration of ML Workflows

Review

high-signal topic

A ride-sharing company built a fraud-detection model that must score each payment transaction within 100 ms while the app is processing the checkout flow, with steady traffic of roughly 500 requests per second throughout the day. Which SageMaker inference option meets these requirements?

Correct answer

B. A SageMaker real-time (persistent) endpoint

A persistent real-time endpoint keeps instances warm behind a low-latency HTTPS endpoint, which is exactly what's needed for sub-second scoring during a live checkout flow with sustained, predictable traffic.

Read full explanation Retake diagnostic

Want the full score report?

Take the free diagnostic, get your weak areas, and save the detailed study plan behind signup.

Start the free diagnostic