ML-Ops

ml-ops.org AI Infrastructure & MLOps Publication

This site explains machine learning operations, commonly shortened to MLOps.

Visit ML-Ops

About ML-Ops

ML-Ops is a web publication that explains machine learning operations. It covers motivation, definition and evolution of MLOps, business requirements for ML-powered software, the end-to-end ML workflow lifecycle, and the roles of data, models, and code.

It also describes principles for testing, deploying, managing, and monitoring models in production, along with CRISP-ML(Q), the MLOps Stack Canvas for specifying architecture and infrastructure stacks, and ML model governance.

Key features

Explains motivation, definition, and evolution of MLOps
Covers business problems and requirements for ML-powered software
Provides overview of end-to-end ML workflow lifecycle
Explains data pipelines, ML workflows, and model serving patterns
Describes principles to test, deploy, manage, monitor models
Covers CRISP-ML(Q), Stack Canvas, and model governance

Best for: Data scientists, ML engineers, and AI enthusiasts learning MLOps concepts.

Good for

Covers topics from business requirements to governance
Describes language-, framework-, platform-, and infrastructure-agnostic practices
Includes application- and industry-neutral stack specification framework