Which activity is part of MLOps lifecycle management?

• A. Building models in isolation without deployment.
• B. Running training only on local machines.
• C. Painting dashboards for visualization.
• D. Automating model lifecycle from development to deployment and monitoring.

Answer: (D)

The essential idea here is that MLOps lifecycle management involves automating the whole journey of a model—from development through deployment to ongoing monitoring. When you automate this end-to-end process, you create repeatable, auditable pipelines that handle data preparation, model training and evaluation, versioning, deployment to production, and continuous monitoring for performance and drift. This enables rapid, reliable updates to models while keeping governance and reproducibility intact.

Why this option is the best fit: automating the entire lifecycle ensures that once a model is developed, it can be consistently trained, tested, deployed, and observed in production without manual handoffs. It supports continuous integration and continuous deployment (CI/CD) for ML, model registries and version control, and automated monitoring and retraining triggers. These elements together constitute the lifecycle management that keeps models healthy and aligned with business needs.

The other activities describe useful tasks but don’t cover the full lifecycle. Building models in isolation without deployment misses the deployment, monitoring, and governance aspects. Running training only on local machines focuses on a single environment, not the scalable, repeatable process needed in production. Painting dashboards for visualization is about presenting results, not managing the lifecycle of models from development to operation.