AI0-001 Practice Test 2 — 30 Questions

Question 1

A data scientist needs to deploy a PyTorch model to production with low-latency inference. The model must be served as a REST API and should support GPU acceleration. Which combination of tools is MOST suitable for this task?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Docker container with a FastAPI application and Nvidia GPU support. It combines Docker containerization with FastAPI for a lightweight REST API and NVIDIA GPU support (via nvidia-docker or NVIDIA Container Toolkit) to enable low-latency GPU-accelerated inference. This stack directly meets the requirements of low-latency inference, REST API serving, and GPU acceleration without unnecessary overhead.

Answer

ONNX runtime with a gRPC endpoint on a CPU-only node

Answer

Apache Spark with MLlib to serve the model in batch mode

Answer

Kubeflow Pipelines to deploy the model as a scheduled job

Question 2

A deep learning engineer is training a convolutional neural network for image classification. The model is overfitting the training data. Which three techniques can help reduce overfitting? (Choose three.)

Accepted Answer

Add dropout layers. Dropout layers randomly deactivate a fraction of neurons during training, which prevents co-adaptation and forces the network to learn more robust features. This reduces overfitting by acting as a form of ensemble learning without increasing model complexity.

Answer

Use a smaller learning rate

Answer

Increase the number of convolutional layers

Question 3

A data scientist is training a binary classification model to detect fraudulent transactions. The dataset contains 99.9% legitimate transactions and 0.1% fraudulent transactions. After training a logistic regression model, the accuracy is 99.9%, but the recall for the fraud class is 0%. Which of the following is the MOST likely cause?

Accepted Answer

The dataset is highly imbalanced, and the model predicts the majority class for all instances.. The dataset has a severe class imbalance (99.9% legitimate, 0.1% fraudulent). A logistic regression model that predicts the majority class (legitimate) for every instance will achieve 99.9% accuracy but 0% recall for the fraud class, because it never identifies any positive fraud cases. This is the classic 'accuracy paradox' in imbalanced classification.

Answer

The regularization parameter is too large, causing underfitting.

Answer

The model is overfitting due to too many features.

Answer

The learning rate was too high.

Question 4

A data scientist is building a binary classification model to predict customer churn. The dataset has 10,000 samples with 80% non-churn and 20% churn. The model achieves 95% accuracy but fails to identify churners correctly. Which metric should the scientist focus on to evaluate model performance properly?

Accepted Answer

Recall (TPR). Recall (True Positive Rate) measures the proportion of actual churners correctly identified by the model. With 80% non-churn and 20% churn, a model can achieve 95% accuracy by simply predicting the majority class (non-churn) for all samples, resulting in zero true positives for churn. Recall directly exposes this failure by quantifying how many churners are captured, making it the critical metric for imbalanced classification problems.

Answer

Precision

Answer

F1-score

Answer

Specificity

Question 5

A company deploys a machine learning model that makes predictions on streaming data. Over time, the data distribution shifts, causing model performance to degrade. Which monitoring strategy is most appropriate to detect this drift?

Accepted Answer

Track the model's accuracy on a fixed validation set over time. Tracking the model's accuracy on a fixed validation set over time directly measures performance degradation caused by data drift. As the streaming data distribution shifts, the model's predictions on the static validation set will become less accurate, providing a clear signal that drift has occurred. This is a standard monitoring approach in production ML systems for detecting concept drift.

Answer

Compare the distribution of predictions to the training set

Answer

Monitor the model's training loss

Answer

Retrain the model daily on new data

Question 6

A team trained a deep neural network on a limited dataset. The training loss decreases consistently, but the validation loss starts increasing after 20 epochs. What is the most likely issue and the best corrective action?

Accepted Answer

Overfitting; apply regularization like dropout. The training loss decreasing while validation loss increasing after 20 epochs is the classic signature of overfitting: the model has memorized the training data but fails to generalize to unseen data. Applying regularization like dropout forces the network to learn more robust features by randomly dropping neurons during training, reducing overfitting. This is the most direct and effective corrective action for this specific symptom.

Answer

Vanishing gradient; use ReLU activation

Answer

Underfitting; increase model complexity

Answer

Data leakage; reshuffle split

Question 7

Which open-source framework is commonly used for building, training, and deploying machine learning models and provides high-level APIs like Keras?

Accepted Answer

TensorFlow. TensorFlow is the correct answer because it is the open-source framework that provides high-level APIs like Keras for building, training, and deploying machine learning models. Keras, now integrated as tf.keras, offers a user-friendly interface for rapid prototyping while TensorFlow handles the underlying computation graph, distributed training, and model serving via TensorFlow Serving.

Answer

Hugging Face Transformers

Answer

scikit-learn

Answer

PyTorch

Question 8

A data scientist is training a binary classification model to detect fraudulent transactions. The dataset is highly imbalanced with only 1% fraud cases. Which technique is most appropriate to address the class imbalance?

Accepted Answer

Oversample the minority class. Oversampling the minority class (e.g., using SMOTE or random oversampling) is the most appropriate technique because it balances the dataset by generating synthetic or duplicate examples of the fraud cases, allowing the model to learn the decision boundary for the minority class without discarding valuable majority-class data. This directly addresses the class imbalance where only 1% of transactions are fraudulent, improving recall and precision for fraud detection.

Answer

Use a linear regression model

Answer

Undersample the majority class

Answer

Increase the learning rate

Question 9

A data engineering team is building a pipeline to ingest streaming user activity data, process it in real-time, and store features in a feature store for ML models. Which streaming technology is BEST suited for this real-time data ingestion and processing?

Accepted Answer

Apache Kafka. Apache Kafka is the best choice because it is a distributed streaming platform designed for high-throughput, fault-tolerant, real-time data ingestion and processing. It provides publish-subscribe messaging, durable log storage, and stream processing capabilities, making it ideal for ingesting streaming user activity data and feeding it into a feature store for ML models.

Answer

Apache Spark SQL

Answer

Apache Airflow

Answer

Apache Hadoop MapReduce

Question 10

A data science team wants to implement a feature store to serve pre-computed features for both training and inference with low latency. Which TWO tools are commonly used for building a feature store?

Accepted Answer

Feast. Feast (Feature Store) is an open-source operational data system that manages and serves machine learning features to both training and inference pipelines with low latency. It provides a consistent feature definition API, offline serving for training, and online serving via a low-latency store like Redis or DynamoDB, making it a standard choice for feature store implementations.

Answer

Kubeflow

Answer

Apache Hive

Answer

MLflow

Question 11

A model trained on a dataset has high bias and low variance. What does this indicate?

Accepted Answer

Underfitting. High bias and low variance indicate that the model is too simple to capture the underlying patterns in the data, leading to systematic errors on both training and test sets. This is the classic signature of underfitting, where the model fails to learn the training data adequately.

Answer

Good fit

Answer

Data leakage

Answer

Overfitting

Question 12

A machine learning team is splitting a dataset for a binary classification problem. They want to ensure robust evaluation and avoid data leakage. Which TWO practices should they follow? (Choose 2)

Accepted Answer

Split into training, validation, and test sets. Option B is correct because splitting the dataset into separate training, validation, and test sets allows the model to be tuned on the validation set while the test set remains untouched for an unbiased final evaluation, which is essential for robust assessment of a binary classifier. Option D is correct because shuffling the data before splitting ensures that the training, validation, and test sets are representative of the overall distribution and prevents ordering bias or temporal artifacts from skewing the split, which helps avoid leakage when the data is not inherently ordered. Option A is not correct because normalising the entire dataset before splitting leaks information from the validation and test sets into the training process, since the scaling parameters would be computed using all data. Option C is not correct because including validation data in the training set removes the independent validation signal needed for hyperparameter tuning and model selection, undermining robust evaluation. Option E is not correct because reusing the same split for all experiments can lead to overfitting to that particular split and does not provide the variability needed to assess generalization reliably.

Answer

Normalise the entire dataset before splitting

Answer

Include validation data in the training set for more data

Answer

Use the same split for all experiments

Question 13

A model trained on customer reviews achieves 98% accuracy on the test set. However, when deployed, it performs poorly on real-world data. The data scientist suspects distribution shift. Which action is MOST important to address this?

Accepted Answer

Implement a monitoring system to detect data drift and retrain with fresh data. Distribution shift (data drift) causes the model's training distribution to differ from the real-world distribution, degrading performance despite high test accuracy. Implementing a monitoring system to detect drift and retraining with fresh data directly addresses this by ensuring the model adapts to the current data distribution, which is the most critical action for maintaining performance in production.

Answer

Reduce the learning rate during training

Answer

Add more features to the model

Answer

Increase the number of cross-validation folds

Question 14

A data scientist needs to store large volumes of unstructured log data for future AI model training. They also need to run SQL-based analytics on the data. Which THREE services are appropriate for this requirement? (Choose 3)

Accepted Answer

Snowflake. Snowflake is correct because it is a cloud-native data warehouse that supports both structured and semi-structured data (like JSON, Avro, Parquet) via its VARIANT data type, enabling SQL-based analytics on unstructured log data. It also integrates with cloud storage (e.g., Amazon S3) for storing large volumes of raw logs, making it suitable for AI model training pipelines.

Answer

Pinecone

Answer

pgvector

Question 15

An ML engineering team has a retraining pipeline that triggers automatically when model accuracy drops below a threshold. Recently, the model's accuracy has been fluctuating, causing frequent retraining and high compute costs. The team suspects the data distribution is changing slowly. Which approach should the team implement to reduce unnecessary retraining while maintaining model performance?

Accepted Answer

Implement a statistical drift detection method on input features. Implementing a statistical drift detection method (e.g., using KL divergence, PSI, or ADWIN) on input features allows the team to identify when the data distribution has genuinely changed, rather than reacting to random accuracy fluctuations. This reduces unnecessary retraining by triggering the pipeline only when statistically significant drift is detected, maintaining model performance without the high compute costs of frequent retraining.

Answer

Use a simpler model to reduce variability

Answer

Increase the frequency of model retraining

Answer

Reduce the batch size for inference

Question 16

Which THREE techniques can help reduce overfitting in neural networks?

Accepted Answer

Increasing training data size. Increasing the training data size helps reduce overfitting by providing the model with more examples to learn from, which reduces the variance and improves generalization. With more data, the model is less likely to memorize noise and instead learns the underlying patterns, making it more robust on unseen data.

Answer

Using a larger learning rate

Answer

Increasing number of layers

Question 17

A team is deploying a generative AI model for a real-time customer-facing application. They need to balance cost and latency. Which deployment strategy is MOST suitable?

Accepted Answer

AI microservices with streaming responses and async processing queues. AI microservices with streaming responses and async processing queues decouple inference from the request lifecycle, allowing the system to handle variable loads efficiently while maintaining low latency for real-time interactions. This architecture balances cost by scaling only the necessary components (e.g., GPU-backed inference services) and uses streaming (e.g., Server-Sent Events or WebSockets) to deliver partial results, reducing perceived latency for the customer.

Answer

Monolithic API with serverless functions

Answer

Edge deployment on user devices

Answer

Batch processing with synchronous requests

Question 18

An organization has a dataset with categorical features having high cardinality (e.g., ZIP codes). They plan to use a tree-based model. Which encoding method is most appropriate?

Accepted Answer

Target encoding (mean encoding). Target encoding (mean encoding) replaces each category with the mean of the target variable for that category, which works well with tree-based models on high-cardinality features because it captures the predictive signal without exploding the feature space. This method avoids the dimensionality explosion of one-hot encoding and the arbitrary ordering of label encoding, making it the most appropriate choice for high-cardinality categorical features in tree-based models.

Answer

Label encoding

Answer

One-hot encoding

Answer

Frequency encoding

Question 19

A company is deploying a generative AI system that produces text content. To comply with emerging transparency obligations, which THREE measures should they implement?

Accepted Answer

Watermark AI-generated content. Watermarking AI-generated content (Option A) is correct because it embeds an imperceptible, machine-detectable signal into the output, enabling provenance verification. This directly addresses transparency obligations by allowing downstream systems to identify synthetic text, which is a key requirement in emerging AI regulations such as the EU AI Act.

Answer

Encrypt all training data

Answer

Limit model access to internal employees only

Question 20

An MLOps team uses a CI/CD pipeline to automate model retraining. The pipeline triggers on new labeled data, runs feature engineering, retrains the model, evaluates against a holdout set, and deploys if metrics exceed thresholds. Recently, a retrained model passed validation but caused a 5% accuracy drop in production. Which improvement best prevents this?

Accepted Answer

Implement canary deployment with shadow scoring to compare with current model. Canary deployment with shadow scoring allows the new model to serve predictions to a small subset of traffic while comparing its outputs against the current production model in real time, without affecting all users. This catches subtle data drift or concept drift that a static holdout set may miss, preventing the 5% accuracy drop from reaching full production.

Answer

Require manual approval before deployment

Answer

Use the entire production dataset for validation instead of a holdout set

Answer

Increase the amount of training data used in each retraining cycle