Get Your CertMLA-C01PlayCompete
‹ MLA-C01 tests

AWS Certified Machine Learning Engineer – Associate (MLA-C01) practice questions

100 questions in 10 tests. Open one to answer it on its own, or take a whole test.

Test 1

Take Test 1 ›
  1. A data science team runs a pandas and scikit-learn feature engineering script by hand inside an Amazon SageMaker Studio notebook. The raw dataset in Amazon S3 has grown to… · Data Preparation
  2. A fraud detection model is served from a real-time endpoint that must look up roughly 40 customer features by customer ID in single-digit milliseconds. The same features are… · Data Preparation
  3. A churn model trained on a tabular customer dataset scores 0.98 AUC on a held-out validation split that was drawn randomly from the same dataset. After deployment, the endpoint's… · Data Preparation
  4. A payments company trains a binary classifier on transactions in which 0.4% of records are fraudulent. The SageMaker AI hyperparameter tuning job is currently configured with… · Model Development
  5. An ML engineer is tuning a PyTorch image classifier that emits a validation:accuracy metric after every epoch. A full training run takes about four hours on the chosen GPU… · Model Development
  6. A SaaS provider trains a separate XGBoost model for each of its 4,000 tenants. Every model artifact is about 30 MB and uses the same serving container. Most tenants send only a… · Deployment and Orchestration
  7. A media company runs a video-analysis model whose input files are roughly 900 MB each and whose inference takes between 8 and 12 minutes per request. Requests arrive… · Deployment and Orchestration
  8. A regulated insurer retrains a pricing model weekly with a SageMaker AI pipeline that processes data, trains, and evaluates the candidate. Compliance requires that a named… · Deployment and Orchestration
  9. A credit-risk model is served from a real-time endpoint. The true outcome for any given application, whether the borrower defaulted, is not known until at least 60 days after the… · Monitoring and Security
  10. A healthcare company trains models on de-identified patient records stored in Amazon S3. The training container is built from a public base image and installs third-party Python… · Monitoring and Security

Test 2

Take Test 2 ›
  1. A lender is preparing a tabular loan dataset in Amazon S3 for a default-prediction model. The annual_income column is missing for about 6% of rows, the gaps are scattered across… · Data Preparation
  2. An ML engineer is preparing features for a linear model that predicts the value of an online order. One input column, payment_method, contains one of four strings: card, wallet… · Data Preparation
  3. A PyTorch training script on SageMaker AI reads a 2 TB image dataset stored as about 20,000 shard files under a single Amazon S3 prefix. The script opens the files by local path… · Data Preparation
  4. A research team retrains a custom PyTorch model a few times each month. Every run takes about 30 hours on GPU instances and is currently billed at On-Demand rates. The team wants… · Model Development
  5. An ML engineer is fine-tuning a 30-billion-parameter transformer on SageMaker AI with PyTorch DistributedDataParallel (DDP) across two instances, each with eight 80 GB GPUs. The… · Model Development
  6. An ML engineer trains a SageMaker AI XGBoost model with max_depth set to 12 and num_round set to 2000. The training logs show training AUC climbing steadily to 0.99. Validation… · Model Development
  7. A bank serves a credit-limit model from a SageMaker AI real-time endpoint and needs to roll out a retrained version. The rollout policy says the new version must first answer a… · Deployment and Orchestration
  8. A retailer hosts a GPU-based recommendation model on a SageMaker AI real-time endpoint. Load testing showed that one instance handles about 400 invocations per minute within the… · Deployment and Orchestration
  9. SageMaker AI training jobs that ran successfully for months now fail while downloading input data, with an AccessDenied error from Amazon S3. The only recent change is that the… · Monitoring and Security
  10. Three data science teams share one AWS account (not part of an organization) for SageMaker AI training jobs, processing jobs, and endpoints. Finance wants to see each team's… · Monitoring and Security

Test 3

Take Test 3 ›
  1. Raw clickstream events land in Amazon S3 as gzip-free CSV files with 120 columns, registered as a table in the AWS Glue Data Catalog. Data scientists build training datasets by… · Data Preparation
  2. Analysts prepare a marketing training dataset in AWS Glue DataBrew, without writing code. One column holds customer email addresses. Policy says the prepared dataset must not let… · Data Preparation
  3. An ML engineer clusters customers with k-means using two numeric features: annual_spend, which ranges from about 200 to 500,000, and orders_per_year, which ranges from 0 to 50… · Data Preparation
  4. A clinical dataset has only 3,000 labeled rows and cannot be expanded. An ML engineer holds out a random 20% of the rows for validation. Re-running the same training code with a… · Model Development
  5. Six ML engineers each run dozens of training jobs a week, some from SageMaker Studio notebooks and some from scripts on their laptops. Hyperparameters and metrics are pasted into… · Model Development
  6. A completed SageMaker AI hyperparameter tuning job ran 100 training jobs and found a best validation RMSE the team wants to improve on. The next tuning job will use the same… · Model Development
  7. A scikit-learn preprocessor fitted during training (imputation values, a scaler, and a category encoder) is currently re-implemented inside three client applications, which… · Deployment and Orchestration
  8. A propensity model scores about 40 million customer records once a night. The records are read from Amazon S3 and the scores are written back to S3 for a marketing system to pick… · Deployment and Orchestration
  9. Users of an application in eu-west-1 report that predictions take about 1.2 seconds at the 99th percentile, against a 400 ms target. The SageMaker AI real-time endpoint is in… · Monitoring and Security
  10. A production SageMaker AI endpoint vanished overnight and inference calls began failing. The security team needs to know which IAM principal called DeleteEndpoint, from what IP… · Monitoring and Security

Test 4

Take Test 4 ›
  1. A payments company scores every card transaction with a fraud model hosted on a SageMaker AI real-time endpoint. The model needs a feature called txn_count_5min, the number of… · Data Preparation
  2. An insurance company plans to train a claims-triage model on documents that several business units have dumped into about 40 Amazon S3 buckets over the years. The files are a mix… · Data Preparation
  3. A retailer is building a customer lifetime value model from customer records merged out of three acquired companies. Many customers appear more than once, but the copies never… · Data Preparation
  4. A utility collects per-minute readings from thousands of smart meters, with about 20 numeric features per reading. It wants to flag unusual readings, such as sudden spikes or… · Model Development
  5. A support team uses a large foundation model on Amazon Bedrock to classify and route incoming tickets. Accuracy is excellent, but at the current volume the model is too expensive… · Model Development
  6. An ML engineer retrained a recommendation model and moved it to a newer inference container. Offline metrics look better, but the team has been burned before by models whose… · Model Development
  7. An engineer packages a custom model server as a Docker image, pushes it to Amazon ECR, and creates a SageMaker AI real-time endpoint from it. After several minutes the endpoint… · Deployment and Orchestration
  8. An HR team uses an internal tool that calls a small scikit-learn model (about 300 MB) to suggest a job category for each new posting. Recruiters use it a few dozen times a day… · Deployment and Orchestration
  9. A web application running on Amazon ECS needs to call a single SageMaker AI real-time endpoint named churn-prod. The account hosts a dozen other endpoints, some of which serve… · Monitoring and Security
  10. A team registered a new model version in the SageMaker Model Registry and must choose the lowest-cost instance type and count for its real-time endpoint. The product requirements… · Monitoring and Security

Test 5

Take Test 5 ›
  1. An AWS Glue ETL job builds the daily training dataset for a demand-forecasting model. Twice last quarter an upstream system shipped a bad extract, once with unit_price empty for… · Data Preparation
  2. A bank is training a binary classifier to flag money-laundering transactions, which make up 0.8% of its 400,000 labeled examples. A data scientist applied SMOTE to the full… · Data Preparation
  3. A medical imaging startup trains its models from a single Amazon S3 bucket in us-east-1. Partner hospitals in Australia, Brazil, and Germany upload 5-20 GB batches of scans to… · Data Preparation
  4. A telecom company stores recordings of its customer support calls in Amazon S3 as stereo audio files, with the agent on one channel and the customer on the other. Managers want a… · Model Development
  5. An engineer runs a SageMaker AI automatic model tuning job for an XGBoost model. Each training job takes about 10 minutes. The tuning job uses the Bayesian strategy over four… · Model Development
  6. A team has a working PyTorch training script that it has been running on an on-premises GPU server. It wants to run the script as SageMaker AI training jobs with as few changes… · Model Development
  7. A platform team manages a CloudFormation stack called ml-network that creates the VPC, private subnets, and security groups for machine learning workloads. Several product teams… · Deployment and Orchestration
  8. An analytics group runs four low-traffic real-time models, each on its own endpoint: a TensorFlow image model, a PyTorch text model, a scikit-learn churn model, and an XGBoost… · Deployment and Orchestration
  9. A company's SageMaker AI spend is steady: real-time endpoints that run around the clock, plus nightly training jobs, together use a consistent baseline of compute every hour… · Monitoring and Security
  10. A payment processor trains a fraud model on cardholder transaction data using a SageMaker AI training job that spans eight GPU instances in private VPC subnets. The training data… · Monitoring and Security

Test 6

Take Test 6 ›
  1. A retail chain streams point-of-sale events into an Amazon Kinesis data stream that feeds a demand-forecasting feature pipeline. The stream is in provisioned mode with 16 shards… · Data Preparation
  2. A hospital network wants to train a model that suggests billing codes from free-text clinical notes. The notes are already stored as UTF-8 text files in Amazon S3. Before the… · Data Preparation
  3. A grocery chain trains a model that predicts next week's sales for each store. The dataset holds five years of weekly rows for 300 stores, sorted by week from oldest to newest… · Data Preparation
  4. A manufacturer wants to detect three kinds of surface defects (scratches, dents, and discoloration) in photos of its metal panels on the production line. Its quality team has… · Model Development
  5. An insurer trains a binary classifier with the SageMaker AI built-in Linear Learner algorithm to flag potentially fraudulent claims for investigation. A regulator requires the… · Model Development
  6. A data scientist trained a scikit-learn model on a laptop and saved it as model.joblib, along with an inference.py script that loads it. To host it, an engineer uploaded both… · Model Development
  7. A SageMaker Pipelines workflow has four steps: a processing step that takes two hours, a training step that takes three hours, an evaluation step, and a model registration step… · Deployment and Orchestration
  8. A bank runs a loan-decision service on Amazon EC2 instances in private subnets. The subnets have no internet gateway or NAT gateway, and security policy forbids adding either… · Deployment and Orchestration
  9. An ML platform team has a monthly allowance of $20,000 for Amazon SageMaker AI. Last month a forgotten GPU endpoint pushed spending over the limit, and finance found out only… · Monitoring and Security
  10. A security engineer must ensure that data scientists can create SageMaker AI training jobs only in approved private subnets. The data scientists' role has one policy that allows… · Monitoring and Security

Test 7

Take Test 7 ›
  1. A subscription company stores customer features, such as support tickets in the last 30 days and logins per week, in a SageMaker Feature Store feature group with the offline… · Data Preparation
  2. A logistics company sends JSON telemetry from delivery vans through Amazon Data Firehose into Amazon S3, where it is later used to train an ETA model. About 3% of events are… · Data Preparation
  3. A data scientist exports a pandas DataFrame with df.to_csv("train.csv", index=False) and uses the file to train the SageMaker AI built-in XGBoost algorithm with the text/csv… · Data Preparation
  4. An online retailer needs to sort product photos into 12 in-house categories. It has 1,500 labeled photos in total, and its data scientists need a model they can tune with their… · Model Development
  5. A software company will use a foundation model on Amazon Bedrock to summarize internal support tickets for engineers. Three candidate models are shortlisted. Before choosing one… · Model Development
  6. A telecom company trains the SageMaker AI built-in XGBoost algorithm with the binary:logistic objective to predict which customers will port their number to a competitor. Only 2%… · Model Development
  7. An ML team deploys a model-serving stack through AWS CodePipeline whenever code merges to main. Its release policy has three rules. Every change must first be deployed to a… · Deployment and Orchestration
  8. A company's data engineering team runs about 200 Apache Airflow DAGs on Amazon Managed Workflows for Apache Airflow (Amazon MWAA). One nightly DAG builds the feature tables for a… · Deployment and Orchestration
  9. A pricing API runs on Amazon API Gateway and an AWS Lambda function. For each request, the function reads customer features from Amazon DynamoDB, calls a SageMaker AI real-time… · Monitoring and Security
  10. A SageMaker AI processing job extracts training data from an Amazon RDS for PostgreSQL database. A security scan found the database password hardcoded in the processing script… · Monitoring and Security

Test 8

Take Test 8 ›
  1. A research lab keeps a 40 TB microscopy image dataset on an on-premises NFS file server, which stays the lab's primary copy. The lab wants to train models on SageMaker AI from… · Data Preparation
  2. A nightly AWS Glue ETL job turns raw JSON clickstream files into Parquet features for a recommendation model. Each day upstream systems add new files under s3://raw/clicks/ and… · Data Preparation
  3. An e-commerce company keeps customer order history in a 2 TB Amazon DynamoDB table that serves production traffic around the clock. Data scientists need a full, consistent… · Data Preparation
  4. A home-goods retailer forecasts weekly demand for 3,000 products. The products follow shared seasonal patterns by category, but many have only a year or two of history. The… · Model Development
  5. A fraud classifier is evaluated on 10,000 validation transactions, 200 of which are actually fraudulent. The results are 80 true positives, 20 false positives, 120 false… · Model Development
  6. A company is building an internal assistant on Amazon Bedrock that answers employee questions about HR policies. About 2,000 policy documents are stored in Amazon S3, and HR… · Model Development
  7. A company hosts six fine-tuned language models, each on its own real-time endpoint backed by an ml.g5.12xlarge instance with four GPUs. The models differ in size: two need all… · Deployment and Orchestration
  8. A lender's credit-scoring model runs on a SageMaker AI real-time endpoint with a single ml.c5.2xlarge instance. Last month an instance failure made the endpoint unavailable for… · Deployment and Orchestration
  9. Right after a new version of a mobile app is released, the SageMaker AI endpoint that powers its product recommendations shows a sharp rise in Invocation4XXErrors… · Monitoring and Security
  10. A company has 30 data science teams sharing one AWS account, and new teams are created every month. Every data scientist's IAM role carries a team tag. The security team requires… · Monitoring and Security

Test 9

Take Test 9 ›
  1. A factory is preparing temperature sensor data in Amazon SageMaker Canvas to train a predictive-maintenance model. The sensors are rated for -40°C to 125°C, but firmware glitches… · Data Preparation
  2. A bank's data lake stores customer tables as Parquet files in Amazon S3, cataloged in the AWS Glue Data Catalog and governed by AWS Lake Formation. Data scientists query the… · Data Preparation
  3. An AWS Lambda function reads an Amazon Kinesis data stream through an event source mapping, enriches each event, and writes features to SageMaker Feature Store one record at a… · Data Preparation
  4. An ad-tech company predicts whether a user will click an ad. Each impression is one-hot encoded from user ID (8 million values), ad ID (300,000 values), and publisher site… · Model Development
  5. An agriculture company flies drones over fields and wants to estimate what percentage of each image is covered by weeds, crops, and bare soil, so it can plan targeted herbicide… · Model Development
  6. A marketing team uses the SageMaker Autopilot API to build a baseline model for a tabular lead-conversion dataset of 45 MB. An engineer explicitly set the job's training mode to… · Model Development
  7. A payroll company's tax-classification endpoint uses a target tracking policy on InvocationsPerInstance. Every weekday at 09:00, when client offices open, traffic jumps about… · Deployment and Orchestration
  8. An ML team has a SageMaker Pipelines pipeline that retrains a demand model on the latest week of data, evaluates it, and registers it. The pipeline must run automatically every… · Deployment and Orchestration
  9. A startup's new AWS account will fine-tune a model on eight ml.p4d.24xlarge instances. Every CreateTrainingJob call fails right away with a ResourceLimitExceeded error saying the… · Monitoring and Security
  10. A healthcare company's auditors require continuous evidence that every SageMaker AI endpoint configuration in its accounts specifies an AWS KMS key for encrypting endpoint… · Monitoring and Security

Test 10

Take Test 10 ›
  1. A company stores 300 TB of historical training datasets in Amazon S3 Standard. Access is unpredictable: a dataset can go untouched for months, then be read in full when a team… · Data Preparation
  2. A smart-home company delivers JSON device events to Amazon S3 through Amazon Data Firehose. Data scientists build training features with Amazon Athena queries that use 6 of each… · Data Preparation
  3. A computer-vision team stores 60 million small training images in one S3 bucket, all directly under the prefix images/ (for example images/000000123.jpg). When several large… · Data Preparation
  4. An insurance company receives thousands of customer emails a day and wants to route each one to one of eight internal queues, such as claims, billing, and policy changes. It has… · Model Development
  5. A news publisher has an archive of 2 million articles with no subject labels. The editors want to discover about 50 recurring themes across the archive and give each article soft… · Model Development
  6. An online bank wants to add a risk score to each customer login so that unusual logins trigger step-up authentication. The main signal is whether a customer is logging in from an… · Model Development
  7. An ML platform team deploys 20 models, each needing a similar SageMaker AI model, endpoint configuration, endpoint, IAM role, and auto scaling setup. The team's engineers write… · Deployment and Orchestration
  8. A travel company's fare-prediction model runs on CPU but needs about 8 GB of memory at inference because it loads large lookup tables alongside the model. It is called a few… · Deployment and Orchestration
  9. A FinOps team suspects that data science teams across 12 AWS accounts have left SageMaker AI real-time endpoints running after experiments ended. The team wants a regularly… · Monitoring and Security
  10. An ML team wants an email within about a minute whenever any SageMaker AI training job in its account ends with a Failed status, whether a pipeline, a notebook, or a scheduled… · Monitoring and Security