People / Students / Class of 2027
Yufan Cheng graduated from the University of Wisconsin–Madison with academic concentrations in Mathematics, Economics, and Statistics. During her undergraduate studies, she built a strong interdisciplinary foundation in quantitative analysis, statistical modeling, machine learning, and data-driven decision-making. Her coursework included big data systems, mathematical methods in data science, artificial intelligence, probability theory, programming, game theory, economic analytics, microeconomic theory, stochastic processes, and deep learning. Through this training, she developed technical proficiency in Python, R, Java, MATLAB, Swift, and HTML.
Prior to joining the MLDS program at Northwestern University, Yufan gained experience across quantitative finance, artificial intelligence, data engineering, and applied research. As an Investment Manager Intern at CUFE Rising Capital Management, she implemented quantitative investment strategies using Python, Pandas, and Random Forest models to support portfolio optimization. She also analyzed financial time-series datasets, applied forecasting methods, and used Monte Carlo simulations to assess risk and identify market trends. Previously, as a Data Analyst Intern at Zhuoshi Quant, she conducted backtesting analyses using financial data from sources such as Barra and Wind, while automating data processing and visualization workflows through Python scripting. Her earlier experience at China Galaxy Securities strengthened her understanding of financial documentation, regulatory compliance, and capital markets.
Yufan has also developed hands-on experience in artificial intelligence and large-scale data systems. As a Core AI/ML Research Intern at DP Technology, she trained the Qwen3 large language model using LoRA and supervised fine-tuning techniques within the LLaMA-Factory framework. She also designed and executed systematic evaluations across multi-domain benchmarks to assess model accuracy, robustness, and efficiency. In her big data systems projects, she built fault-tolerant data services and streaming pipelines using technologies such as gRPC, Cassandra, Spark, Kafka, HDFS, Protobuf, PyArrow, Google Cloud Storage, BigQuery, and Dataform. These projects strengthened her ability to connect machine learning workflows with scalable data infrastructure.
Her research experiences reflect a broader interest in applying data science to complex real-world systems. As a Research Assistant on a robotics industry data analysis project supervised by Prof. Qiongjie Zheng of the Chinese Academy of Engineering and Nanjing University, she structured high-noise qualitative datasets from robotics firms, applied importance–feasibility analysis to identify industry-wide bottlenecks, and co-authored a report for the Chinese Academy of Engineering. At the Lab for System Informatics and Data Analytics, supervised by Prof. Kaibo Liu, she built a Python framework for generating spatiotemporal sensor data streams with Gaussian-process-style fields, localized shifts, and missingness patterns to benchmark anomaly detection methods. She also led a predictive analysis project on medically underserved areas, merging over 13,000 regional records from HRSA and ACS, using correlation analysis and Random Forest models to identify socioeconomic predictors, and building a linear regression model to forecast MUA/P designations.
Through the MLDS program at Northwestern University, Yufan aims to deepen her technical expertise in machine learning, statistical modeling, and data systems while continuing to explore how data-driven methods can be applied to finance, AI research, robotics, and public-policy challenges. She is particularly interested in the intersection of machine learning, quantitative decision-making, and real-world deployment, where technical models must be interpretable, reliable, and connected to practical institutional needs. She looks forward to contributing to a collaborative and interdisciplinary learning environment, bringing together her background in mathematics, economics, statistics, finance, and AI to solve complex problems with both analytical rigor and practical impact.
