# What is MLOps? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Model Operations  
> **Canonical Citation:** [https://spidits.com/ai-glossary/mlops](https://spidits.com/ai-glossary/mlops)

## Definition
MLOps (Machine Learning Operations) is a set of practices, culture, and tools focused on automating and unifying the lifecycle of machine learning models, spanning data collection, training, testing, deployment, and monitoring.

## Architectural Mechanics & Deep Dive
MLOps (Machine Learning Operations) is a set of practices that aims to deploy, monitor, and maintain machine learning models in production reliably and efficiently. Combining software engineering, data engineering, and DevOps, MLOps automates model packaging, versioning, deployment testing, drift monitoring, and periodic retraining pipelines.

## Industry Applications & Real-World Use Cases
Automated CI/CD model pipelines, prediction monitoring dashboards, and model registry management.

## Frequently Asked Questions

### How does MLOps differ from DevOps?
DevOps focuses on code versioning and software deployment. MLOps deals with three axes: code versioning, dataset versioning, and model weight parameters, which are non-deterministic and decay over time.

### What is model drift in MLOps?
Model drift occurs when the statistical properties of production input data change over time relative to the training data, leading to a decay in prediction accuracy and requiring retraining.

## Related Terminology
- [model-registry](https://spidits.com/ai-glossary/model-registry)
- [observability](https://spidits.com/ai-glossary/observability)
- [inference](https://spidits.com/ai-glossary/inference)

---
*From the SPIDITS AI Knowledge Base (300+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*