Cross-Domain Recommendation is a new eld of study in the area of recommender systems. The goal of these recommender systems is to use information from other domains to provide recommendations in target domains. In this work, we focus on content-based cross-domain recommendation, specifically job recommendation. Instead of defining the notion of domain based on item descriptions, we introduce user-based domains. We dene meta-data features, such as current job function of users and current user industry, as domain indicators. We introduce \indicator features" to segment users into different domains and propose an efficient method of feature augmentation to implement domain adaptation among various meta-data features. The logistic regression algorithm used in this work is chosen due to its simplicity of implementation, flexibility of utilizing different kinds of features, and extensibility. We experimented based on both online metrics, such as precision, recall, and accuracy, and online metrics using A/B testing. The results show the effectiveness of our cross-domain content-based method in a real-world study.