Published teaching material

Making ML Engineering Practical

An end-to-end machine learning engineering curriculum connecting source control, testing, data systems, modeling, APIs, and containers.

2019–present · Instructor and curriculum designer
TeachingMachine LearningInfrastructure
Title slide for the BDA-602 Machine Learning Engineering course taught by Dr. Julien Pierret
Contribution boundary

I designed and taught the course material. The examples and public notes support students learning practical ML engineering rather than representing a production application.

Why it exists

Machine learning education often spends most of its time on models and very little on the engineering required to make those models repeatable, testable, deployable, and understandable by another person.

I designed this curriculum around the full path from an idea to a maintained system.

What I teach

  • Git and collaborative source control
  • Testing, linting, and Python package structure
  • SQL, Spark, and large-scale data processing
  • Feature engineering and model evaluation
  • FastAPI-based model services
  • Docker and reproducible runtime environments
  • Later Big Data and Generative AI material

Teaching as engineering evidence

The project demonstrates more than presentation skills. Building a coherent curriculum requires choosing useful abstractions, sequencing dependencies, writing examples that survive close inspection, and explaining tradeoffs to people with different levels of experience.

The public material is useful both as a course archive and as evidence of how I communicate practical engineering decisions.