#490: How Amazon Ads Run Machine Learning at Scale

Amazon Ads operates its ad serving infrastructure at massive scale, handling hundreds of millions of ad requests per second (trillions of ads per day) within a latency budget of 120 milliseconds. They use machine learning to predict relevant ads to show shoppers, predict whether shoppers will click or purchase, and allocate and price an ad. In this episode, Simon talks with Kun Liu, Director of Machine Learning at Amazon Ads, about how his team uses machine learning running on AWS to score thousands of ads per request within a 20ms window at over 100K transactions per second, all across hundreds of models that each required different hardware and software optimizations. Learn more about Amazon Ads - https://advertising.amazon.com/ Read the blog - https://go.aws/31rPqM3 Learn more about AWS solutions for advertising & marketing - https://aws.amazon.com/advertising-marketing/