Google Professional Machine Learning Engineer Exam Dumps and Practice Test Questions Set 13 Q181-195

Visit here for our full Google Professional Machine Learning Engineer exam dumps and practice test questions.

Question 181

A bank wants to implement real-time credit scoring for thousands of loan applications per second. The system must provide low-latency decisions, continuously adapt to new financial behaviors, and scale with growing application volume. Which architecture is most appropriate?

A) Batch process applications nightly and manually evaluate credit scores
B) Use Pub/Sub for real-time application ingestion, Dataflow for feature computation, and Vertex AI Prediction for online scoring
C) Store applications in spreadsheets and manually compute scores
D) Train a model once on historical data and deploy it permanently

Answer: B

Explanation:

Batch processing applications nightly and manually evaluating credit scores is inadequate for real-time decision-making. Credit applications require instantaneous decisions, and nightly processing introduces unacceptable delays, which can frustrate customers and reduce operational efficiency. Manual evaluation cannot scale to thousands of applications per second and is prone to errors. Batch workflows also do not support continuous retraining, preventing the system from adapting to evolving financial behaviors, economic changes, or shifting risk profiles. This approach fails to meet the operational and regulatory requirements of modern banking systems.

Using Pub/Sub for real-time application ingestion, Dataflow for feature computation, and Vertex AI Prediction for online scoring is the most appropriate solution. Pub/Sub ingests applications in real time, ensuring no data is lost. Dataflow pipelines continuously compute features such as debt-to-income ratios, payment history, credit utilization, and behavioral patterns, which are critical inputs for scoring models. Vertex AI Prediction provides low-latency credit scoring, enabling immediate decisions on loan approvals or denials. Continuous retraining pipelines allow models to adapt to changes in customer behavior, emerging financial risks, and updated regulations, maintaining predictive accuracy. Autoscaling ensures the system can handle peak application volumes efficiently. Logging, monitoring, and reproducibility provide operational reliability, auditability, and compliance. This architecture ensures scalable, low-latency, and continuously adaptive credit scoring for operational use.

Storing applications in spreadsheets and manually computing credit scores is impractical. Spreadsheets cannot efficiently handle thousands of real-time applications, and manual computation is slow, error-prone, and non-reproducible.

Training a model once on historical data and deploying it permanently is insufficient. Customer behavior, economic conditions, and regulatory requirements evolve continuously. Static models cannot adapt, resulting in decreased predictive accuracy and potential financial risk. Continuous retraining and online scoring are necessary to maintain operational effectiveness.

The optimal solution is Pub/Sub for real-time ingestion, Dataflow for feature computation, and Vertex AI Prediction for online scoring, providing scalable, low-latency, and continuously adaptive credit decision-making.

Question 182

A telecommunications company wants to detect anomalies in network traffic in real time. The system must scale to millions of log entries per second, provide low-latency alerts, and continuously adapt to new patterns of network failures. Which architecture is most appropriate?

A) Batch process logs nightly and manually inspects anomalies
B) Use Pub/Sub for log ingestion, Dataflow for feature computation, and Vertex AI Prediction for online anomaly detection
C) Store logs in spreadsheets and manually identify anomalies
D) Train a model once on historical logs and deploy it permanently

Answer: B

Explanation:

Batch processing logs nightly and manually inspecting anomalies is inadequate for real-time network anomaly detection. Network conditions can change rapidly, and nightly batch processing introduces unacceptable delays, preventing timely detection of issues that can impact service quality. Manual inspection cannot scale to millions of log entries per second and is error-prone. Batch workflows do not support continuous retraining, which is essential for adapting to evolving failure patterns and new anomalies. This approach is unsuitable for modern telecommunications operations that require operational real-time monitoring and immediate mitigation.

Using Pub/Sub for log ingestion, Dataflow for feature computation, and Vertex AI Prediction for online anomaly detection is the most appropriate solution. Pub/Sub ingests device logs in real time, ensuring all events are captured. Dataflow pipelines continuously compute derived features such as latency spikes, packet loss, error rates, and correlations across multiple devices, which are critical for detecting anomalies. Vertex AI Prediction provides low-latency anomaly detection, enabling immediate alerts and automated mitigation actions. Continuous retraining pipelines allow models to adapt to emerging failure patterns, maintaining predictive accuracy. Autoscaling ensures efficient handling of high-volume log streams. Logging, monitoring, and reproducibility provide operational reliability, traceability, and compliance with industry regulations. This architecture supports scalable, low-latency, and continuously adaptive network anomaly detection.

Storing logs in spreadsheets and manually identifying anomalies is impractical. Spreadsheets cannot efficiently process millions of entries, and manual computation is slow, error-prone, and non-reproducible.

Training a model once on historical logs and deploying it permanently is insufficient. Network behavior evolves, and static models cannot detect new anomalies, reducing detection accuracy and operational reliability. Continuous retraining and online computation are necessary to maintain effectiveness.

The optimal solution is Pub/Sub for real-time ingestion, Dataflow for feature computation, and Vertex AI Prediction for online anomaly detection, providing scalable, low-latency, and continuously adaptive monitoring.

Question 183

A retailer wants to forecast demand for hundreds of products across multiple stores using historical sales, promotions, holidays, and weather data. The system must scale to millions of records, allow feature reuse, and continuously update forecasts. Which solution is most appropriate?

A) Train separate models locally for each store using spreadsheets
B) Use Vertex AI Feature Store for centralized features and Vertex AI Training for distributed forecasting
C) Store historical sales data in Cloud SQL and train a single global linear regression model
D) Use a simple rule-based system based on last year’s sales

Answer: B

Explanation:

Training separate models locally for each store using spreadsheets is impractical. Retail datasets include millions of records across hundreds of stores and multiple products, covering features such as historical sales, promotions, holidays, and weather. Local training cannot efficiently process this volume, and manual workflows are slow, error-prone, and non-reproducible. Managing features separately for each store introduces redundancy and inconsistency. Automated retraining pipelines are difficult to implement locally, and feature reuse is limited. This approach is unsuitable for enterprise-level demand forecasting that requires scalability, automation, and accuracy.

Using Vertex AI Feature Store for centralized features and Vertex AI Training for distributed forecasting is the most appropriate solution. Feature Store ensures consistent, reusable features across multiple models, reducing duplication and ensuring consistency between training and serving. Vertex AI Training supports distributed training across GPUs or TPUs, efficiently processing millions of historical records while capturing complex patterns in sales, promotions, holidays, and weather. Automated pipelines handle feature updates, retraining, and model versioning, ensuring forecasts continuously improve as new data becomes available. Autoscaling ensures efficient processing for large datasets. Logging, monitoring, and experiment tracking provide reproducibility, operational reliability, and governance compliance. This architecture enables scalable, accurate, and continuously updated inventory demand forecasts across multiple stores and products.

Storing historical sales data in Cloud SQL and training a single global linear regression model is insufficient. Cloud SQL is not optimized for large-scale analytical workloads, and linear regression cannot capture non-linear patterns. A single global model may underfit, producing inaccurate forecasts and lacking localized precision.

Using a simple rule-based system based on last year’s sales is inadequate. Rule-based approaches cannot account for promotions, holidays, weather, or changing trends. They lack scalability, automation, and predictive accuracy, making them unsuitable for enterprise-level forecasting.

The optimal solution is Vertex AI Feature Store combined with Vertex AI Training, providing scalable, reusable, and continuously updated inventory demand forecasts.

Question 184

A bank wants to detect fraudulent credit card transactions in real time for millions of users. The system must scale to high transaction volumes, provide low-latency scoring, and continuously adapt to new fraud patterns. Which architecture is most appropriate?

A) Batch process transactions daily and manually review suspicious activity
B) Use Pub/Sub for transaction ingestion, Dataflow for feature computation, and Vertex AI Prediction for online fraud scoring
C) Store transactions in spreadsheets and manually compute fraud risk
D) Train a model once per year and deploy it permanently

Answer: B

Explanation:

Batch processing transactions daily and manually reviewing suspicious activity is inadequate for real-time fraud detection. Credit card fraud occurs quickly, and daily batch processing introduces unacceptable delays, leaving fraudulent transactions undetected. Manual review cannot scale to millions of transactions efficiently and is prone to human error. Batch workflows also prevent continuous retraining, limiting the model’s ability to adapt to emerging fraud patterns, reducing predictive accuracy. This approach does not meet the operational and regulatory requirements for modern banking systems that require immediate fraud detection and mitigation.

Using Pub/Sub for transaction ingestion, Dataflow for feature computation, and Vertex AI Prediction for online fraud scoring is the most appropriate solution. Pub/Sub enables high-throughput, real-time ingestion of transaction data, ensuring that no transactions are missed. Dataflow pipelines compute derived features such as transaction frequency, location anomalies, spending patterns, and device behavior. Vertex AI Prediction provides low-latency scoring, allowing immediate action on suspicious transactions. Continuous retraining pipelines ensure models adapt to evolving fraud tactics, maintaining predictive accuracy over time. Autoscaling guarantees efficient processing during peak transaction periods. Logging, monitoring, and reproducibility ensure operational reliability, auditability, and regulatory compliance. This architecture provides scalable, low-latency, and continuously adaptive fraud detection, supporting operational resilience and real-time mitigation.

Storing transactions in spreadsheets and manually computing fraud risk is impractical. Spreadsheets cannot efficiently process high-volume transaction data, and manual computation is slow, error-prone, and non-reproducible, making it unsuitable for operational monitoring.

Training a model once per year and deploying it permanently is insufficient. Fraud patterns evolve rapidly, and static models cannot detect new behaviors, reducing accuracy and increasing financial risk. Continuous retraining and online scoring are essential for maintaining operational effectiveness.

The optimal solution is Pub/Sub for real-time ingestion, Dataflow for feature computation, and Vertex AI Prediction for online fraud scoring, providing scalable, low-latency, and continuously adaptive transaction monitoring.

Question 185

A logistics company wants to optimize delivery routes dynamically using real-time vehicle telemetry, traffic data, and weather conditions. The system must scale to thousands of vehicles, provide low-latency route recommendations, and continuously adapt to changing conditions. Which solution is most appropriate?

A) Batch process delivery and traffic data daily, and manually update routes
B) Use Pub/Sub for real-time data ingestion, Dataflow for feature computation, and Vertex AI Prediction for online routing optimization
C) Store vehicle and traffic data in spreadsheets and manually compute optimal routes
D) Train a route optimization model once and deploy it permanently

Answer: B

Explanation:

Batch processing, delivering, and traffic data daily, and manually updating routes daily is insufficient for real-time route optimization. Traffic congestion, vehicle availability, and weather conditions change frequently, and daily updates result in outdated routes. Manual computation cannot scale to thousands of vehicles and is prone to human error. Batch workflows also lack continuous retraining, preventing models from adapting to evolving conditions, which reduces predictive accuracy. This approach is unsuitable for enterprise logistics requiring low-latency, dynamic route optimization.

Using Pub/Sub for real-time data ingestion, Dataflow for feature computation, and Vertex AI Prediction for online routing optimization is the most appropriate solution. Pub/Sub allows continuous ingestion of vehicle telemetry, traffic updates, and weather information, ensuring all events are captured in real time. Dataflow pipelines compute derived features such as expected delays, congestion impact, and vehicle availability, providing critical inputs for accurate route recommendations. Vertex AI Prediction delivers low-latency online routing recommendations, enabling dispatch systems to respond immediately. Continuous retraining pipelines allow models to adapt to evolving traffic patterns, vehicle behavior, and environmental factors, maintaining accurate predictions. Autoscaling ensures efficient processing during peak delivery times. Logging, monitoring, and reproducibility provide operational reliability, traceability, and governance compliance. This architecture supports scalable, low-latency, and continuously adaptive route optimization.

Storing vehicle and traffic data in spreadsheets and manually computing optimal routes is impractical. Spreadsheets cannot handle high-frequency telemetry efficiently. Manual computation is slow, error-prone, non-reproducible, and unsuitable for continuous operational routing.

Training a route optimization model once and deploying it permanently is insufficient. Traffic, vehicle availability, and environmental conditions evolve continuously, and static models cannot adapt. Continuous retraining and online computation are necessary to maintain accurate routing recommendations.

The optimal solution is Pub/Sub for real-time ingestion, Dataflow for feature computation, and Vertex AI Prediction for online routing optimization, providing scalable, low-latency, and continuously adaptive routing.

Question 186

A retailer wants to forecast inventory demand across multiple stores using historical sales, promotions, holidays, and weather conditions. The system must scale to millions of records, allow feature reuse, and continuously update forecasts. Which solution is most appropriate?

Answer: B

Explanation:

Training separate models locally for each store using spreadsheets is impractical. Retail datasets include millions of records across hundreds of stores and multiple products, encompassing historical sales, promotions, holidays, and weather data. Local training cannot efficiently process this volume and is slow, error-prone, and non-reproducible. Managing features separately for each store introduces redundancy and inconsistency. Automated retraining pipelines are difficult to implement locally, and feature reuse is limited. This approach is unsuitable for enterprise-level demand forecasting that requires scalability, automation, and accuracy.

Storing historical sales data in Cloud SQL and training a single global linear regression model is insufficient. Cloud SQL is not optimized for large-scale analytical workloads, and linear regression cannot capture complex nonlinear relationships. A single global model may underfit, producing inaccurate forecasts and lacking localized precision.

The optimal solution is Vertex AI Feature Store combined with Vertex AI Training, providing scalable, reusable, and continuously updated inventory demand forecasts.

Question 187

A healthcare provider wants to predict patient readmission using EHR data, lab results, and imaging. The system must comply with privacy regulations, scale to large datasets, and support reproducible training pipelines. Which solution is most appropriate?

A) Download all patient data locally and train models manually
B) Use BigQuery for structured data, Cloud Storage for unstructured data, and Vertex AI Pipelines for preprocessing, training, and deployment
C) Store patient data in spreadsheets and manually compute readmission risk
D) Train a model once using sample data and deploy it permanently

Answer: B

Explanation:

Downloading all patient data locally and training models manually is not suitable due to privacy, compliance, and scalability issues. EHR data is highly sensitive and regulated under HIPAA and other healthcare standards. Local storage increases the risk of unauthorized access, and manual workflows are slow, error-prone, and non-reproducible. Local computation cannot efficiently process heterogeneous datasets like structured lab results and unstructured imaging. Manual preprocessing can introduce inconsistencies, and retraining pipelines are difficult to implement, limiting adaptability to new patient data or evolving clinical guidelines. This approach fails to meet operational and regulatory requirements for secure healthcare predictive analytics.

Using BigQuery for structured data, Cloud Storage for unstructured data, and Vertex AI Pipelines for preprocessing, training, and deployment is the most appropriate solution. BigQuery efficiently manages structured EHR data, enabling scalable queries for patient demographics, lab results, medication history, and vitals. Cloud Storage securely handles unstructured data, including imaging and clinical notes, ensuring scalable access while maintaining privacy compliance. Vertex AI Pipelines orchestrates preprocessing, feature extraction, model training, and deployment in a reproducible manner, ensuring consistency and traceability. Continuous retraining allows models to adapt to new patient information, evolving treatments, and updated clinical protocols, maintaining predictive accuracy. Logging, monitoring, and experiment tracking ensure operational reliability, auditability, and regulatory compliance. Autoscaling enables the processing of growing datasets efficiently. This architecture provides secure, scalable, reproducible, and continuously adaptive readmission risk prediction.

Storing patient data in spreadsheets and manually computing readmission risk is impractical. Spreadsheets cannot handle large-scale structured and unstructured healthcare datasets, and manual computation is slow, error-prone, and non-reproducible, making it unsuitable for operational prediction.

Training a model once using sample data and deploying it permanently is insufficient. Patient populations, treatments, and clinical practices evolve, and static models cannot adapt to new trends, resulting in decreased predictive accuracy. Continuous retraining and automated pipelines are essential for operational effectiveness.

The optimal solution is BigQuery for structured data, Cloud Storage for unstructured data, and Vertex AI Pipelines for secure, scalable, reproducible, and continuously adaptive readmission risk prediction.

Question 188

A logistics company wants to optimize delivery routes dynamically using real-time vehicle telemetry, traffic, and weather data. The system must scale to thousands of vehicles, provide low-latency route recommendations, and continuously adapt to changing conditions. Which approach is most appropriate?

A) Batch process delivery and traffic data daily, and manually update routes
B) Use Pub/Sub for real-time data ingestion, Dataflow for feature computation, and Vertex AI Prediction for online route optimization
C) Store vehicle and traffic data in spreadsheets and manually compute optimal routes
D) Train a route optimization model once and deploy it permanently

Answer: B

Explanation:

Batch processing, delivery, and traffic data daily and manually updating routes is insufficient for real-time route optimization. Traffic conditions, vehicle availability, and weather change constantly. Daily batch updates result in outdated routing, reducing delivery efficiency and customer satisfaction. Manual computation cannot scale to thousands of vehicles and is error-prone. Batch workflows also lack continuous retraining, preventing adaptation to changing patterns, which reduces predictive accuracy. This approach is unsuitable for enterprise logistics operations requiring low-latency, dynamic routing optimization.

Using Pub/Sub for real-time data ingestion, Dataflow for feature computation, and Vertex AI Prediction for online route optimization is the most appropriate solution. Pub/Sub continuously ingests vehicle telemetry, traffic updates, and weather information, ensuring real-time event capture. Dataflow pipelines compute derived features such as congestion impact, expected delays, and vehicle availability, providing critical inputs for accurate route recommendations. Vertex AI Prediction delivers low-latency routing recommendations, enabling dispatch systems to respond immediately. Continuous retraining allows models to adapt to evolving traffic patterns, vehicle behavior, and environmental factors. Autoscaling ensures efficient processing during peak periods. Logging, monitoring, and reproducibility provide operational reliability, traceability, and compliance. This architecture ensures scalable, low-latency, and continuously adaptive route optimization.

Storing vehicle and traffic data in spreadsheets and manually computing optimal routes is impractical. Spreadsheets cannot efficiently process high-frequency telemetry. Manual computation is slow, error-prone, non-reproducible, and unsuitable for continuous operational routing.

Training a route optimization model once and deploying it permanently is insufficient. Traffic patterns, vehicle availability, and environmental conditions evolve continuously. Static models cannot adapt, reducing predictive accuracy. Continuous retraining and online computation are necessary to maintain efficient routing.

The optimal solution is Pub/Sub for ingestion, Dataflow for feature computation, and Vertex AI Prediction for online route optimization, providing scalable, low-latency, and continuously adaptive routing.

Question 189

Answer: B

Explanation:

Using Vertex AI Feature Store for centralized features and Vertex AI Training for distributed forecasting is the most appropriate solution. Feature Store ensures consistent, reusable features across multiple models, reducing duplication and maintaining training-serving consistency. Vertex AI Training supports distributed training across GPUs or TPUs, efficiently processing millions of historical records while capturing complex patterns in sales, promotions, holidays, and weather. Automated pipelines handle feature updates, retraining, and model versioning, ensuring forecasts continuously improve as new data becomes available. Autoscaling ensures efficient processing of large datasets. Logging, monitoring, and experiment tracking provide reproducibility, operational reliability, and governance compliance. This architecture enables scalable, accurate, and continuously updated inventory demand forecasts across multiple stores and products.

Storing historical sales data in Cloud SQL and training a single global linear regression model is insufficient. Cloud SQL is not optimized for large-scale analytical workloads, and linear regression cannot capture complex non-linear relationships. A single global model may underfit, producing inaccurate forecasts and lacking localized precision.

The optimal solution is Vertex AI Feature Store combined with Vertex AI Training, providing scalable, reusable, and continuously updated inventory demand forecasts.

Question 190

Answer: B

Explanation:

Batch processing applications nightly and manually evaluating credit scores is inadequate for real-time decision-making. Credit applications require immediate approval or rejection, and nightly batch processing introduces delays that negatively impact customer experience and operational efficiency. Manual evaluation cannot scale to thousands of applications per second and is prone to errors. Batch workflows do not support continuous retraining, preventing the system from adapting to new financial behaviors, emerging economic conditions, or evolving risk profiles. This approach fails to meet operational and regulatory requirements for modern banking.

Using Pub/Sub for real-time application ingestion, Dataflow for feature computation, and Vertex AI Prediction for online scoring is the most appropriate solution. Pub/Sub ensures that all applications are ingested in real time without loss. Dataflow pipelines compute features such as debt-to-income ratios, payment history, credit utilization, and behavioral patterns critical for accurate credit scoring. Vertex AI Prediction provides low-latency scoring for immediate decisions. Continuous retraining pipelines allow models to adapt to changes in customer behavior, regulatory updates, and new financial risks, maintaining predictive accuracy. Autoscaling ensures the system handles peak application volumes efficiently. Logging, monitoring, and reproducibility provide operational reliability, auditability, and compliance. This architecture ensures scalable, low-latency, and continuously adaptive credit scoring.

Storing applications in spreadsheets and manually computing scores is impractical. Spreadsheets cannot handle high-frequency, high-volume data efficiently. Manual computation is slow, error-prone, and non-reproducible, making it unsuitable for operational environments.

Training a model once on historical data and deploying permanently is insufficient. Customer behavior, economic conditions, and regulations evolve continuously. Static models cannot adapt, reducing accuracy and increasing risk. Continuous retraining and online scoring are necessary for operational effectiveness.

Question 191

A telecommunications company wants to detect anomalies in network traffic in real time. The system must scale to millions of log entries per second, provide low-latency alerts, and continuously adapt to evolving network failure patterns. Which architecture is most appropriate?

A) Batch process logs nightly and manually inspect anomalies
B) Use Pub/Sub for log ingestion, Dataflow for feature computation, and Vertex AI Prediction for online anomaly detection
C) Store logs in spreadsheets and manually identify anomalies
D) Train a model once on historical logs and deploy permanently

Answer: B

Explanation:

Batch processing logs nightly and manually inspecting anomalies is inadequate for real-time detection. Network conditions can deteriorate rapidly, and nightly batch processing introduces delays that prevent timely detection, potentially causing downtime or degraded service. Manual inspection cannot scale to millions of log entries per second and is prone to errors. Batch workflows do not allow continuous retraining, limiting the system’s ability to adapt to new patterns of network failures. This approach is unsuitable for modern telecommunications operations that require real-time monitoring and immediate mitigation.

Using Pub/Sub for log ingestion, Dataflow for feature computation, and Vertex AI Prediction for online anomaly detection is the most appropriate solution. Pub/Sub enables high-throughput, real-time ingestion of network logs without loss. Dataflow pipelines continuously compute derived features, such as latency spikes, packet loss, error rates, and correlations across devices, which are critical for detecting anomalies. Vertex AI Prediction provides low-latency online anomaly detection, allowing immediate alerts and automated mitigation actions. Continuous retraining allows models to adapt to evolving network behaviors, maintaining predictive accuracy. Autoscaling ensures efficient handling of high-volume log streams. Logging, monitoring, and reproducibility provide operational reliability, traceability, and compliance. This architecture ensures scalable, low-latency, and continuously adaptive network monitoring.

Storing logs in spreadsheets and manually identifying anomalies is impractical. Spreadsheets cannot process millions of records efficiently, and manual analysis is slow, error-prone, and non-reproducible.

Training a model once on historical logs and deploying permanently is insufficient. Network traffic evolves constantly, and static models cannot detect new patterns, reducing anomaly detection accuracy. Continuous retraining and online computation are necessary for effective real-time monitoring.

The optimal solution is Pub/Sub for log ingestion, Dataflow for feature computation, and Vertex AI Prediction for online anomaly detection, providing scalable, low-latency, and continuously adaptive network monitoring.

Question 192

A retailer wants to forecast inventory demand across hundreds of products in multiple stores using historical sales, promotions, holidays, and weather. The system must scale to millions of records, support feature reuse, and continuously update forecasts. Which solution is most appropriate?

Answer: B

Explanation:

Training separate models locally for each store using spreadsheets is impractical. Retail datasets include millions of records across multiple stores and products, covering historical sales, promotions, holidays, and weather data. Local training cannot efficiently process this volume and is slow, error-prone, and non-reproducible. Managing features separately for each store introduces redundancy and inconsistency. Automated retraining pipelines are difficult to implement locally, and feature reuse is limited. This approach is unsuitable for enterprise-level demand forecasting, which requires scalability, automation, and predictive accuracy.

Using Vertex AI Feature Store for centralized features and Vertex AI Training for distributed forecasting is the most appropriate solution. Feature Store ensures consistent, reusable features across multiple models, reducing duplication and ensuring training-serving consistency. Vertex AI Training supports distributed training across GPUs or TPUs, efficiently processing millions of historical records while capturing complex patterns in sales, promotions, holidays, and weather. Automated pipelines handle feature updates, retraining, and model versioning, ensuring forecasts continuously improve as new data becomes available. Autoscaling ensures efficient processing for large datasets. Logging, monitoring, and experiment tracking provide reproducibility, operational reliability, and governance compliance. This architecture enables scalable, accurate, and continuously updated inventory demand forecasts across multiple stores and products.

The optimal solution is Vertex AI Feature Store combined with Vertex AI Training, providing scalable, reusable, and continuously updated inventory demand forecasts.

Question 193

A bank wants to detect fraudulent credit card transactions in real time for millions of users. The system must scale to high transaction volumes, provide low-latency scoring, and continuously adapt to emerging fraud patterns. Which architecture is most appropriate?

Answer: B

Explanation:

Batch processing transactions daily and manually reviewing suspicious activity is insufficient for real-time fraud detection. Fraud can occur within seconds, and daily batch processing introduces delays that leave fraudulent transactions undetected, increasing financial and reputational risk. Manual review cannot scale to millions of transactions efficiently and is prone to human error. Batch workflows also prevent continuous retraining, which is necessary to adapt to new fraud patterns and evolving user behaviors. This approach does not meet operational requirements for real-time fraud prevention and is unsuitable for modern banking systems.

Using Pub/Sub for transaction ingestion, Dataflow for feature computation, and Vertex AI Prediction for online fraud scoring is the most appropriate solution. Pub/Sub enables high-throughput, real-time ingestion of transactions, ensuring no events are lost. Dataflow pipelines continuously compute derived features such as transaction frequency, location anomalies, device behavior, and spending patterns, which are critical for accurate scoring. Vertex AI Prediction provides low-latency scoring for immediate fraud alerts and interventions. Continuous retraining pipelines ensure models adapt to new fraud tactics and user behavior, maintaining predictive accuracy. Autoscaling guarantees efficient processing during peak transaction volumes. Logging, monitoring, and reproducibility provide operational reliability, auditability, and regulatory compliance. This architecture ensures scalable, low-latency, and continuously adaptive fraud detection.

Storing transactions in spreadsheets and manually computing fraud risk is impractical. Spreadsheets cannot handle high-volume transaction data efficiently, and manual computation is slow, error-prone, and non-reproducible, making it unsuitable for operational monitoring.

Training a model once per year and deploying it permanently is insufficient. Fraud patterns evolve rapidly, and static models cannot detect new behaviors, resulting in decreased predictive accuracy and financial risk. Continuous retraining and online scoring are necessary to maintain operational effectiveness.

Question 194

A logistics company wants to optimize delivery routes dynamically using real-time vehicle telemetry, traffic updates, and weather conditions. The system must scale to thousands of vehicles, provide low-latency recommendations, and continuously adapt to changing conditions. Which solution is most appropriate?

A) Batch process delivery and traffic data daily, and manually update routes
B) Use Pub/Sub for real-time data ingestion, Dataflow for feature computation, and Vertex AI Prediction for online route optimization
C) Store vehicle and traffic data in spreadsheets and manually compute optimal routes
D) Train a route optimization model once and deploy it permanently

Answer: B

Explanation:

Batch processing, delivery, and traffic data daily and manually updating routes is insufficient for real-time route optimization. Traffic congestion, vehicle availability, and weather conditions change frequently. Daily updates result in outdated routes, reducing delivery efficiency and customer satisfaction. Manual computation cannot scale to thousands of vehicles and is prone to human error. Batch workflows do not support continuous retraining, limiting adaptation to changing conditions and reducing predictive accuracy. This approach is unsuitable for enterprise logistics operations requiring low-latency, dynamic route optimization.

Using Pub/Sub for real-time data ingestion, Dataflow for feature computation, and Vertex AI Prediction for online route optimization is the most appropriate solution. Pub/Sub continuously ingests telemetry, traffic, and weather data, ensuring all updates are captured in real time. Dataflow pipelines compute derived features such as expected delays, congestion impact, and vehicle availability, providing critical inputs for accurate route recommendations. Vertex AI Prediction delivers low-latency online route optimization, enabling immediate adjustments in dispatch operations. Continuous retraining pipelines allow models to adapt to evolving traffic patterns, vehicle behavior, and environmental conditions. Autoscaling ensures efficient processing during peak delivery times. Logging, monitoring, and reproducibility provide operational reliability, traceability, and compliance. This architecture ensures scalable, low-latency, and continuously adaptive route optimization.

Storing vehicle and traffic data in spreadsheets and manually computing routes is impractical. Spreadsheets cannot efficiently handle high-frequency telemetry. Manual computation is slow, error-prone, and non-reproducible, making it unsuitable for continuous operational routing.

Training a route optimization model once and deploying it permanently is insufficient. Traffic, vehicle availability, and environmental factors evolve continuously. Static models cannot adapt, reducing routing accuracy. Continuous retraining and online computation are necessary to maintain efficiency.

Question 195

A retailer wants to forecast demand for hundreds of products across multiple stores using historical sales, promotions, holidays, and weather conditions. The system must scale to millions of records, support feature reuse, and continuously update forecasts. Which solution is most appropriate?

Answer: B

Explanation:

Training separate models locally for each store using spreadsheets is impractical. Retail datasets include millions of records across multiple stores and products, covering features such as historical sales, promotions, holidays, and weather. Local training cannot efficiently process this volume, and manual workflows are slow, error-prone, and non-reproducible. Managing features separately for each store introduces redundancy and inconsistency. Automated retraining pipelines are difficult to implement locally, and feature reuse is limited. This approach is unsuitable for enterprise-level demand forecasting, which requires scalability, automation, and predictive accuracy.

Storing historical sales data in Cloud SQL and training a single global linear regression model is insufficient. Cloud SQL is not optimized for large-scale analytical workloads, and linear regression cannot capture complex nonlinear relationships. A single global model may underfit, producing inaccurate forecasts and lacking localized precision.

The optimal solution is Vertex AI Feature Store combined with Vertex AI Training, providing scalable, reusable, and continuously updated inventory demand forecasts.

Google Professional Machine Learning Engineer Exam Dumps and Practice Test Questions Set 13 Q181-195

Related posts: