Xuanjing (Jean) Chen stands out as a professional who bridges academic research and industry innovation in recommendation systems, personalization algorithms, and data science. Her work combines scholarly rigor with real-world impact, reshaping how AI potentially connects people, content, and opportunities at scale. This balance highlights the relevance of her work, showing how technical research can influence digital ecosystems and user engagement worldwide.
Chen’s academic journey has established the foundation for her technical and analytical depth. She earned a Master of Science in Business Statistics (Marketing Science) from Columbia Business School, graduating with a merit scholarship, after completing her Bachelor of Science in Media, Culture, and Communication with a Data Science minor at New York University. Her scholarly contributions include the paper “A Context-Aware Personalized Recommendation Framework Integrating User Clustering and BERT-Based Sentiment Analysis,” which suggests that advanced NLP and clustering methods could significantly improve recommendation accuracy, and “A Machine Learning–Based Enterprise Financial Audit Framework and High-Risk Identification,” which applied models such as SVM, Random Forest, and KNN to enterprise risk detection. These works have been recognized as expanding the frontier of both personalized recommendation and applied AI in business contexts.
Chen has since translated this research expertise into progressive industry leadership roles. At Bigo Live, she applied her data science skill sets to analyze streamers’ performance and user behavior, directly influencing strategy and contributing to a 19% increase in monthly revenue and a 37% rise in user retention. This ability to translate technical complexity into practical business strategies became a defining characteristic of her professional style. At Habu (acquired by LiveRamp), she advanced the application of data clean rooms by embedding machine learning models into marketing analytics pipelines, delivering measurable ROI improvements for global clients such as ASICS. Her role involved refining customer segmentation and attribution methods, which helped marketing teams make more informed decisions and reduce inefficiencies in advertising spending. By focusing on transparency and reproducibility in data workflows, she ensured that clients could reliably evaluate the outcomes of their marketing investments.
Now at TikTok, within AI Data Service & Operations – Search Ecosystem, Chen is driving the next wave of personalization infrastructure. She designed and launched a three-pillar messaging system—comprising personalized newsletters, general CSI marketing, and item-to-user activations—covering more than 50 million subscribers. These innovations, powered by clustering, multimodal modeling, and AI-driven recommendation, led to a 296% DAU growth and a 120% DCC growth QoQ, while resulting in a reduction of operational costs by over 50%. She also built TikTok’s first personalized newsletter recommendation system, praised by creators as an essential tool for inspiration, milestone tracking, and growth. Her approach emphasized scalability and sustainability: the systems were not one-time experiments but frameworks designed to evolve alongside TikTok’s rapidly growing user base. She also carefully documented key processes extensively, which then allowed other teams within the company to adopt and extend her methods, multiplying the impact beyond her immediate projects.
Across academia and industry, Chen has consistently shown the ability to bridge theory and practice—taking complex statistical models and machine learning frameworks from research to deployment at a global scale. Her contributions demonstrate the transformative potential of AI-driven personalization, not only for business growth but for shaping how digital ecosystems can empower creators and audiences worldwide. By demonstrating that academic methods can be adapted to operational environments, her work underscores the importance of linking rigorous research with applied problem-solving. The systems she has developed show that personalization technologies can scale responsibly when designed with both accuracy and efficiency in mind. In this way, her career provides an example of how innovation may move from theoretical exploration into tools that actively shape user experience in global platforms.
