Introduction to fully automated MLOps

Fully Automated MLOps

Arturo Opsetmoen Amador

Senior Consultant - Machine Learning

MLOps in an industrial setting

What are common goals companies have when using machine learning?

  • Develop ML tools and products that use data to
    • Better serve customers
    • Optimize processes

An image showing how a company can use machine learning systems to either enhance internal processes or to improve customer facing services.

Fully Automated MLOps

Optimizing for value generation

Companies aim to maximize profits

  • Machine learning can be used to increase profitability
  • By analyzing costs and revenues, a company can estimate potential profits
  • Deploying the right number of ML use-cases can result in profits

A plot showing a curve. On the x-axis, the number of machine learning projects. On the y-axis the revenue generated by the projects. The curve increases until it reaches a plateau at 6 projects and a revenue of approx. 25 million. After this plateau, the revenue decreases without ever increasing again.

Fully Automated MLOps

Costs in software development

Expected costs in a traditional software development project:

  • Development costs
  • Project management
  • UI/UX design
  • Quality assurance
Fully Automated MLOps

Technical debt in software development

Technical debt or design debt:

  • Cost of rework caused by poor design

Picture of two engineers in front of a poorly built house. One says to the other "I don't understand why it takes so long to add a new window."

1 https://vincentdnl.com/drawings/
Fully Automated MLOps

Hidden technical debt in ML systems

Machine Learning: "The high-interest credit card of technical debt"[1].

Hidden technical debt can be related to:

  1. The data used to train the ML models
  2. The models powering the ML system
  3. The infrastructure used by the ML system
  4. The monitoring of the ML system
1 https://research.google/pubs/pub43146/
Fully Automated MLOps

Costs of machine learning projects

Reduced profit due to technical debt in ML systems: $$ $$

Image showing two figures with one curve each. In the first figure, an estimated revenue of $25M is calculated. In the second figure, this revenue has been reduced to $10M.

Fully Automated MLOps

The high-interest credit card of technical debt

ML systems can be complex and become unruly.

A picture showing the many components present in an ML system. ML code is a small component at the center. Around it we have configuration, data collection, data validation, machine resource management, serving infrastructure, monitoring, analysis tools, process management tools, feature extraction.

Fully Automated MLOps

MLOps: The best-known way to pay

If ML is the high-interest credit card of technical debt, MLOps is the best way to pay for it.

MLOps can include:

  • Automated testing
  • Automated experiment tracking
  • Automated monitoring

To keep the technical debt to a minimum

Fully Automated MLOps

Let's practice!

Fully Automated MLOps

Preparing Video For Download...