What Is MLOps?

MLOps Explainer

MLOps is a set of engineering practices for building and operating machine-learning systems reliably. It extends ideas from DevOps while accounting for data, experiments, models, evaluation, and changing real-world conditions in addition to application code.

A production workflow may validate data, train models, track artifacts, run tests, compare candidates, approve a version, deploy it, monitor outcomes, and trigger investigation or retraining. Automation reduces manual variation, but human review may remain essential for consequential releases.

Machine-learning systems can fail even when the model file and serving code are unchanged. Input distributions shift, upstream schemas break, labels arrive late, and feedback loops alter behavior. MLOps therefore includes monitoring data quality and model performance over time.

The goal is not maximum tooling. A useful approach matches process to risk and scale, with versioned data and code, reproducible runs, clear ownership, staged deployment, rollback plans, access controls, documentation, and evidence that the deployed system still meets its requirements.

MLOps applies engineering practices to the full machine-learning lifecycle, including data, training, deployment, monitoring, and retraining.

Key terms include machine-learning lifecycle, pipeline, continuous integration, continuous delivery, continuous training, artifact, monitoring, rollback and governance.

Production ML includes data collection, validation, experimentation, deployment, serving, observation, incident response, and retirement.

A reproducible pipeline connects those stages and records which inputs and code produced each model version.

Automated retraining can move quickly, but it can also propagate bad data or an inappropriate objective without effective checks.

Release criteria, approvals, monitoring, rollback, lineage, and ownership should reflect the consequences of failure.

Explore more "Explainers"

Discover additional explainers across politics, science, business, technology, and other fields. Each explainer breaks down a complex idea into clear, everyday language—helping you better understand how major concepts, systems, and debates shape the world around us.