MLflow

MLflow is an open-source platform to manage the end-to-end machine learning lifecycle, including experiment tracking, model packaging, and deployment. It provides lightweight APIs and a tracking server to record runs, artifacts, and model versions. MLflow integrates with common ML frameworks and supports reproducible workflows across teams.

Use this profile to understand the building block briefly, place it in the model, and switch to the 360° assessment when needed.

Content type

Technology

Technical building block: can be automated, integrated, or operated.

Classification level

Element

Concrete cog in the system that works inside larger relationships.

Available perspectives

Open this building block directly in an already curated view.

360°

In brief

MLflow – in brief

MLflow is an open-source platform for making machine-learning experiments, models, and deployment traceable.

Where does it come from?

MLflow was developed at Databricks in 2018 to ease the move from exploratory machine-learning experiments to reproducible workflows, and was then released as an open-source project. It grew into a platform for tracking, model packaging, registry, and deployment.

The mental model

A training run records parameters, metrics, artifacts, and code. A registry entry versions a model and can represent transitions through deployment stages. Serving or integrations load that version while teams compare and roll back models.

Key building blocks

Run

A run groups inputs, measurements, and results from a training or evaluation execution.

Model registry

The registry manages model versions and their lifecycle.

Model serving

Serving exposes a registered model for predictions through an interface.

When is it useful?

MLflow provides traceability between experiments and operations and fits heterogeneous ML stacks. Teams still control tracking content, access, data and model risks, and actual production execution; the platform does not guarantee model quality.

MLflow documentation MLflow – GitHub repository MLflow – Wikipedia (unavailable in audit: HTTP 404)