# MLOps

Concept · [Building with AI](https://ecosystem.samcarlton.com/ai)

Also called: ML ops, LLMOps, machine learning operations, AI ops

The practices and tools for running [machine learning models](https://ecosystem.samcarlton.com/model) in [production](https://ecosystem.samcarlton.com/production): versioning data and models, deploying them, [monitoring](https://ecosystem.samcarlton.com/observability) quality and retraining when needed.

**For example:** An MLOps setup [alerts](https://ecosystem.samcarlton.com/observability) the team when a fraud model's accuracy drops and retrains it on recent data.

## Connects to (4)

- [DevOps](https://ecosystem.samcarlton.com/devops): The practice, and job, of combining software development with running it in production: automating builds, tests and deployments and…
- [Evals](https://ecosystem.samcarlton.com/evals): Tests that check whether the outputs of a model or an AI product meet the criteria its builders set.
- [Observability](https://ecosystem.samcarlton.com/observability): Being able to see what a running system is doing through its logs, metrics and traces, so a team can tell when something goes wrong and why.
- [Training](https://ecosystem.samcarlton.com/training): Teaching an AI model by showing it a very large number of examples before anyone uses it.

Map view: https://ecosystem.samcarlton.com/mlops