Fighting against Marketplace Scams using Community-driven AI

Today's fast-moving world of physical collectibles and digital assets / NFTs is fraught with scams designed to drain people out of their hard-earned money. Those looking to buy or simply talk about the collectibles and brands they love are often left to fend for themselves armed only with hard-to-parse documentation about best practices. A machine learning pipeline that can protect users from bad actors in real time while continuously learning new malicious behaviors from community members needs to constantly learn and keep track of features in a highly-available feature store. This talk will cover how Shiba uses a Feature Store to promote inter-team collaboration and maintain machine learning models with up-to-date data on how bad actors target communities.