Learn about how Pinterest reduced the iOS app size, how to measure dev team's efficiency, how to use graph neural networks at scale, how 150M images are being searched at NYT and how LinkedIn streamlined search for 44% of the users.

Welcome to the 8th issue of engineering brew. 

If you are an EM, and believe in measuring different aspects of your team’s efficiency - this is a great read. Author talks about the importance, avoidable mistakes and shares three effective ways to measure a group’s efficiency. 

https://www.cloudbees.com/blog/measuring-engineering-efficiency/?ref=engineeringbrew.com

Author Liang Ma discussed the problem of iOS app size, and why it was important to reduce the size of the app. I love that not only the fix was discussed here, but also how the team will keep doing this as part of their long term plan.

https://medium.com/pinterest-engineering/how-we-reduced-pinterests-ios-app-size-by-30-50mb-68d7f8425882/?ref=engineeringbrew.com

This is a take on how the team re-architected a 20 year old system that now has more than 150 million photos. 

https://open.nytimes.com/over-150-million-photos-some-perl-scripts-and-a-creaky-user-interface-c00f7b6d5fd9/?ref=engineeringbrew.com

A streamlined search can be a super-power for your user. And re-imagining it always takes a lot of buy-in, guts and effort. In this post LinkedIn engineering team talks about their approach, the rewrites and the results of building a streamlined search experience.     

https://engineering.linkedin.com/blog/2021/building-a-more-intuitive-and-streamlined-search-experience?ref=engineeringbrew.com

This post briefly explains neural networks, shares one of the reasons why they are not used in more applications and then really digs into how to solve that problem. Or I should say, how they solved it at Twitter. 

https://blog.twitter.com/engineering/en_us/topics/insights/2021/simple-scalable-graph-neural-networks.html?ref=engineeringbrew.com