The Gang of Four book on design patterns remains a cornerstone of software engineering, guiding developers to write cleaner, more maintainable, and scalable code. As a data engineer with a software engineering background, I’ve often asked myself: why don’t we have a similar concept of design patterns in data engineering?

Over the past three years, I’ve made it a practice to try to identify and apply reusable patterns while building data pipelines, searching for ways to make ETLs more maintainable, DRY, and production-ready. These patterns not only save time and reduce complexity but also bring essential structure to a field that often leans on ad hoc solutions.

In this talk, I’ll introduce the concept of data engineering design patterns, what they are, why they matter, and how to apply them in real-world pipelines. I’ll begin by outlining common categories of patterns, then take a closer look at two critical areas: data ingestion and data storage. For each, I’ll present key patterns, describe the problems they address, explain the design principles behind them, and explore implementation variations from the field.

This session is ideal for data engineers, software developers transitioning into data roles, or anyone looking to bring software engineering concepts into their data infrastructure.