How Strava built an algorithmically-backed discovery feed in 7 weeks

Last year, Strava wanted to provide an easy way for our athletes to be inspired by the vast amount of content the Strava community around the world produces. My team was tasked with creating a surface that suggests content, like routes, segments and clubs, to our athletes. This surface needed to be personalized and generated algorithmically. With a team of 6 engineers, we shipped this new product in 7 weeks. In this talk, I will tell the story of how we accomplished this and the technologies we utilized to achieve our goals and build a scalable product.