MLflow Foundation

}
2 Days

Available On-Site

Available Virtually

Contact Us for Open Enrollment
f

Customizable

MLflow Foundation is a two-day course that helps you take machine learning workflows from ad-hoc scripts to production-grade pipelines. Day one opens with a framework overview, then shows how to organize experimental work into Projects. Projects are reproducible units that bundle code, dependencies, and entry points, so a teammate can rerun an experiment exactly as it was built. Attendees package models in a standard format that downstream tools can load regardless of which library trained them. They also track each run’s parameters, metrics, and artifacts so results stay comparable across dozens or hundreds of experiments. Hands-on labs reinforce each concept, giving attendees practice logging their own runs before the class moves on.

Day two turns to utilities and advanced management. Participants register trained models in the Model Registry, moving them through versions and lifecycle stages such as staging and production as a model matures. From there, attendees deploy registered models to a serving endpoint for real-time predictions and explore batch inference for scoring larger datasets. The class finishes with the command-line interface and the web GUI. Attendees use both to search, filter, and compare experiments side by side across a project. They can also rerun a prior experiment from its recorded parameters.

Participants leave with an in-depth understanding of the platform’s architecture, deployment options, and day-to-day usage. They also gain hands-on experience turning their own workflows into production-ready processes.

Who Should Attend

Developers, Data Scientists, Data Engineers, ML Engineers, DevOps & AI/MLOps professionals

What Attendees Will Learn

Upon completing MLflow Foundation, participants will be able to:

  • Explain MLflow’s architecture, deployment options, and use
  • Configure and manage MLflow and its components
  • Organize machine learning workflows as Projects and Runs
  • Package and track models using MLflow
  • Deploy models using the MLflow Model Registry
  • Use the MLflow CLI and GUI to manage and search experiments
  • Transform machine learning workflows into production grade, end-to-end solutions

Prerequisites

Participants should have basic programming skills (preferably in Python) and solid command-line skills.

Delivery

Available for Instructor-Led (ILT) in-person/onsite training or Virtual Instructor-Led training (VILT) delivery.

Each attendee will require the ability to ssh into a cloud hosted virtual machine (provided with the course). In environments where SSH is not possible, local lab VMs or browser accessible lab systems can be provided. For web-based delivery, participants require an Internet-connected computer capable of teleconferencing.

Related Instructor-Led (ILT & VILT) Training Courses

If you are interested in other Cloud Native, AI, programming, or other courses, check out the full course list or search our entire catalog:

Secret Link