Yahoo enhanced its Search Retargeting (SRT) capabilities in its omnichannel Demand-Side Platform (DSP) using Amazon Bedrock, addressing challenges in connecting user search intent with relevant ad experiences. The legacy SRT architecture used the Word2Vec embedding model combined with locality-sensitive hashing (LSH) to expand advertiser keywords, but this approach had limitations including outdated vocabulary, phrase-based searches, reliance on syntactic rather than semantic similarity, and occasional failure to generate new keywords. To overcome these issues, Yahoo deployed generative AI–powered SRT keyword expansion using Amazon Bedrock and large language models (LLMs). This enhancement generates more relevant, semantically rich keyword expansions, helping advertisers reach larger and more precise audiences. SRT is a core audience targeting solution that helps advertisers reach users based on historical search behavior, bridging search intent with display, video, and native advertising. Beyond targeting keywords entered on Yahoo Search, SRT uses AI to identify and engage users who demonstrate intent through search activity both on Yahoo and across integrated partner systems. The Yahoo DSP provides advertisers with advanced technology, premium supply access, vast scale, and trusted consumer relationships, allowing ad inventory purchases across multiple exchanges and channels through a single interface. Audience segments, defined as groups of users sharing specific interests, demographics, or behaviors, are passed to targeting systems for ad serving.