O'Reilly logo
live online training icon Live Online training

Deploying machine learning models to production: A toolkit for real-world success

Armen Donigian

Companies are looking for ways to incorporate machine learning into their business to save money and increase revenue. In this hands-on practical training, Armen Donigian walks you through designing, developing, deploying, and monitoring machine learning models in production and shares common pitfalls and best practices to help you get started with your own generalizable machine learning platforms.

What you'll learn-and how you can apply it

  • Machine learning deployment pipeline basic concepts
  • How version control differs for machine learning projects
  • Considerations to take into account with respect to feature engineering, model evaluation, and model interpretability
  • How to manage package dependencies for experimental and critical path work
  • Pitfalls during model development
  • A hardware cost estimation model
  • The differences between offline and online model training
  • The differences between static and dynamic inference

And you’ll be able to:

  • Build generalizable machine learning models
  • Develop and deploy a production-ready machine learning model
  • Estimate the hardware capacity for a given model
  • Optimize trained machine learning models to reduce inference latency
  • Monitor performance of a machine learning model in production

This training course is for you because...

  • You're a data scientist, business analyst, or machine learning engineer who wants to build or deploy generalizable models in production.
  • You're a data engineer or software engineer who wants to deploy machine learning models in production.


  • Intermediate software engineering or data science skills

Required materials and setup:

  • A machine with a modern browser (Chrome, Firefox, etc.) installed
  • Bookmark or download materials from course Github repo (link will be provided closer to the event date)

Recommended preparation:

About your instructor

  • Armen Donigian has undergraduate and graduate degrees in Computer Science from UCLA and USC. He started his career building tracking & navigation algorithms at Orincon (later acquired by Lockheed Martin). Armen then joined the Global Differential GPS group at Jet Propulsion Laboratory (NASA), performing clock and orbit corrections using GPS/GLONASS satellites, which were also used for testing of Mars Science Laboratory Curiosity Rover.

    Bitten by the startup bug, Armen has helped several startups build data driven products and scale infrastructure as a Senior Data & Machine Learning Engineer. Armen currently works as the Director of Data Science Engineering at Zest Finance.


The timeframes are only estimates and may vary according to how the class is progressing

Day 1:

An overview of the machine learning deployment pipeline (20 minutes)

  • Lecture: Introduction to the machine learning deployment pipeline; key questions to ask in an end-to-end data science process; common vocabulary
  • Knowledge check exercise

Break (10 minutes)

How version control differs for machine learning projects (95 minutes)

  • Lecture: A repeatable and reproducible model training, test, and validation pipeline; model interpretability methods; exploratory data analysis using open source tooling
  • Knowledge check exercise

Break (10 minutes)

Package dependencies (10 minutes)

  • Lecture: How and why to manage dependencies, including those by third parties
  • Knowledge check exercise

Cloud hosting cost estimation tools (35 minutes) - Lecture: Various hardware deployment configurations; how to price each offering; Amazon AWS and Google Cloud platform resources
- Knowledge check exercise

Day 2:

Pitfalls during model development (40 minutes)

  • Lecture: The dos and don'ts of building a model for production
  • Programming exercise: Detect common pitfalls such as overfitting, data leakage, and nonstationary data

Break (10 minutes)

Offline versus online model training (50 minutes)

  • Lecture: Use cases and reference implementations for offline and online model training
  • Programming exercise: Implement an improved online learning model

Break (10 minutes)

Offline versus online model inference (40 minutes)

  • Lecture: Use cases and reference implementations for offline and online inference
  • Programming exercise: Implement an improved offline batch scoring model

Model monitoring in production (30 minutes)

  • Lecture: Why and how to monitor a machine learning model in production
  • Programming exercise: Implement an input variable distribution monitor