Grateful for Less Waste: A Thanksgiving Reflection on Sustainability, Data, and the Package Decision Engine

Thanksgiving is a season of gratitude for family, for good food, and for the small, often invisible systems that make modern life work. As I made my comeback to LinkedIn and returned to technical writing about my mission around Data as a Product, I found myself reflecting on where that philosophy started. Much of it traces back to my work on the Packaging Decision Engine, the AI-driven brain that determines the right packaging for every order at Amazon. That model alone has helped Amazon avoid over 4.2 million metric tons of packaging material worldwide since 2015.

That experience shaped how I think about data today. Optimizing packaging is remarkably similar to how we should engineer data systems. Oversized boxes, unnecessary padding, and duplicated materials have their digital equivalents: oversized tables, redundant pipelines, and uncontrolled data growth. Waste is waste, whether physical or digital and both require intentional design to eliminate.

Today, 12% of global Amazon orders ship without any added Amazon packaging. Achievements like this don’t happen by chance. They result from millions of micro-decisions powered by well-structured, well-governed data. And that is exactly why my mission now is to build Data as a Product (DaaP).

Just as PDE aims to ship fewer boxes, DaaP aims to ship less data, but with higher quality, stronger governance, and clearer purpose. Less noise. Less redundancy. Less operational burden. More impact.

This Thanksgiving, I’m grateful for two things: the systems that help us reduce physical waste, and the data principles that help us reduce digital waste.

Both make the world and our work meaningfully better.

Happy Thanksgiving, y’all.