AI engineering glossary · MLOps, LLMOps cloud
What is MLOps?
MLOps: A set of practices that combine machine learning, software engineering and operations to build, deploy, monitor and maintain models reliably in production. It covers versioning, automation, testing, deployment and monitoring so models keep working after the first launch.
MLOps: an example
An MLOps team automates retraining a demand-forecasting model weekly, tests it, deploys it and alerts on accuracy drops.
Related terms
More MLOps, LLMOps cloud terms
- Model artefact
- Model monitoring
- Model promotion
- Model registry
- Model validation
- Model versioning
- Monitoring
- Multi-stage build
Where you learn it
MLOps is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.