Evaluating your ML project with the MLOps checklist - AWS Prescriptive Guidance

Evaluating your ML project with the MLOps checklist

Charles Frenzel, Sharath Nagaraja, and Spencer Romo Amazon Web Services (AWS)

July 2023 (document history)

The MLOps checklist is a workable checklist that you can use at any phase in your machine learning (ML) project. The checklist is a tool for assessing overall readiness, examining system coverage, and identifying new areas of opportunity in distributed ML systems. MLOps is the combination of people, technology, and processes for delivering ML solutions. Well-architected MLOps helps businesses to deploy ML models to production effectively and consistently, and can deliver business value.

Using the MLOps checklist helps you to do the following:

  • Assess your MLOps system.

  • Find areas of opportunity.

  • Find areas for improvement.

  • Evaluate and update your strategic roadmap on AWS.

  • Generate backlog items.

We recommend using the MLOps checklist at the start of your MLOps project, but it's possible to use parts of it during any phase.