The top Pinterest Engineering blog posts of 2020

In 2020, Pinterest engineers launched more shopping features than ever, enabling retailers to bring their products to Pinterest and Pinners to discover and try on the items that work best for them. With the pandemic came the acceleration of online shopping, and Pinterest was ready to serve and scale for merchants, with catalog uploads growing 10x in the first half of the year.

In quarantine, global monthly active users grew 37% year over year to 442 million at the end of the quarter. The team supported and scaled foundational parts of the platform like search, home feed, recommendations and ads, as well as the growth and safety. 2020 also saw the launch of new tools for creators, collaboration features for group boards, and the Today tab to bring daily and expert information to Pinners around the world. And it was all mostly done from home.

Here are the most-read posts of the year, that go behind the scenes of some of that work:

How a one line change decreased our clone times by 99%

How Pinterest Supercharged its Growth Team With Experiment Idea Review

Using graph algorithms to optimize Kafka operations, Part 1

Interest Taxonomy: A knowledge graph management system for content understanding at Pinterest

Using PID controllers to diversify content types on home feed

Launching your career as an apprentice engineer at Pinterest

Deepak Agarwal: My First 90 Days at Pinterest

A better clickthrough rate: How Pinterest upgraded everyone’s favorite engagement metric

Using machine learning to auto-organize boards

Contextual relevance in ads ranking

We’ll be back in the new year to share more stories from behind the technology and teams at Pinterest. Happy Holidays!