An under-the-hood look into how machine learning (ML) decides what goes into people’s News Feeds through the Facebook app. In this project I worked with researchers and engineers to develop infographics while using the AI blog brand to match and the accompany video in explaining how the data is parsed through a closer lens for Engineering @Meta.
Read the accompanying blog.
Coming in late to the project, my role in the video was towards the end to help prep animation files for the movable characters to ensure that animators had the pieces organized to do their job. While the video was in post I was responsible for using the art style to create online assets such as this hero image and the following diagrams and flows.
The purpose of this image is to give a static sharable ‘under the hood’ view to audiences that might want a deeper dive into the research. In order to convey a siphoning-structure in which to describe how the ranking algorithm works, I mirrored the video’s style and added elements of movement and depth, while maintaining a clear systematic view on the tech.
Here is an architectural diagram to help audiences in the weeds such as engineers and research peers and journalists understand the ML further.