AI AI Engineer: MLOps and Model Deployment 1 — Questions and Answers
Question 1: Which MLOps practice ensures that a model retrained on new data maintains or improves its performance compared to the previous version?
- Model versioning
- Model validation and comparison (Correct answer)
- Feature engineering
- Data augmentation
Correct answer: Model validation and comparison
Model validation and comparison ensures newly trained models meet or exceed baseline performance before being promoted to production.
Question 2: What is the primary purpose of a feature store in an MLOps pipeline?
- To store trained model weights
- To provide a centralized repository for reusable, versioned feature data (Correct answer)
- To monitor model drift in production
- To schedule training jobs
Correct answer: To provide a centralized repository for reusable, versioned feature data
A feature store centralizes feature computation and storage so teams can share, reuse, and version features consistently across training and serving.
Question 3: Which deployment strategy releases a new model to a small subset of users before rolling out to everyone?
- Blue-green deployment
- Canary deployment (Correct answer)
- Shadow deployment
- Rolling deployment
Correct answer: Canary deployment
Canary deployment directs a small percentage of traffic to the new model, limiting blast radius if issues arise.
Question 4: What does 'data drift' mean in the context of production ML models?
- Training data has errors
- The statistical distribution of input features changes over time (Correct answer)
- Model weights change during inference
- Feature pipelines run slower
Correct answer: The statistical distribution of input features changes over time
Data drift occurs when the real-world input distribution shifts away from the distribution the model was trained on, degrading performance.
Question 5: Which container orchestration platform is most commonly used to deploy and scale ML model serving workloads?
- Apache Kafka
- Kubernetes (Correct answer)
- Apache Spark
- Airflow
Correct answer: Kubernetes
Kubernetes automates deployment, scaling, and management of containerized applications, making it the standard for production ML serving.
Question 6: In CI/CD for ML, what does a 'model registry' primarily store?
- Raw training datasets
- Versioned trained models with metadata and lineage (Correct answer)
- API keys for cloud services
- Docker container images
Correct answer: Versioned trained models with metadata and lineage
A model registry stores versioned model artifacts along with metadata like metrics, lineage, and stage (staging, production) for governance.
AI AI Engineer: MLOps and Model Deployment 1
Which MLOps practice ensures that a model retrained on new data maintains or improves its performance compared to the previous version?