Skip to main content

Getting Started

Welcome to Michelangelo AI! Whether you're evaluating the platform or ready to build your first ML pipeline, you're in the right place.

Want a single ordered path instead of a menu? The Learning Paths page sequences the docs into three role-based tracks — ML Engineer, Operator, and Contributor — each with time estimates and a clear endpoint.

Choose your path​

I want to understand what Michelangelo AI does​

Start here if you're evaluating the platform or want to learn the key concepts before diving in. Takes about 15 minutes.

  • Overview — What Michelangelo AI is, how it works, and how familiar ML tools map to it
  • Core Concepts and Key Terms — Projects, workflows, tasks, and the key terms you'll encounter
  • MLOps Glossary — Alphabetical reference for all Michelangelo AI terms, plus a concept mapping table for users coming from MLflow, Kubeflow, Ray, or Airflow

I want to set up and start building​

Ready to get hands-on? Follow these guides in order. You'll have a working local environment in about 20 minutes.

  1. Sandbox Setup — Set up a local Michelangelo AI cluster with all services running (~20 min)
  2. Getting Started with Pipelines — Build and run your first ML pipeline (~30 min)
  3. Browse Examples — 10 end-to-end workflows: XGBoost, BERT, GPT fine-tuning, batch inference, and more

Reference​

For contributors​

Building on Michelangelo AI's core platform? These guides cover the contributor development workflow: