New York, New York, United States
I am currently a graduate student major in statistics with an applied mathematics bachelor’s degree which provides me a strong background in understanding the hidden principle of doing data analysis. Also, I have always been a numbers person, with exceptional mathematical and computational skills with fluency in several data analytical software and skills, including Python, R, Java, Power BI, Excel, Tableau, MySQL, Event Tracking, A/B Testing, and Machine Learning.
Built Python-based customer segmentation and funnel analysis across 50+ product categories; applied a Random Forest targeting model that increased customer retention by 37.5% (3.2% to 4.4%) and contributed to a 3.7% ($550K) increase in GMV
• Extract and organize data from a multitude of sources to feed business-improving insights (user engagement, Social Media Analytics, Product Usage Analytics) and optimize business outcomes • Analyze user behavior data to uncover actionable insights which can be used to provide learnings and recommendations to the Client Leadership team • Employ factor analysis (EFA and CFA) on user behavior survey data, pinpointing the top 4 categories which contribute the most influential reasoning to our target value • Collaborate cross-functionally with different teams to enhance data utilization efficiency and craft compelling data narratives that drive strategic decision-making. • Build thoughtful presentations and reports to deliver the right performance narrative to clients • Perform ETL and EDA with large scale image data (up to 2 TB) and perform big data scale statistical test across different regions of interest to identify indicator and testify reliability among different dataset.
• Designed web crawling scripts that automated the downloading and organization of over 800 academic documents towards 4 different Columbia University Academic Commons (AC) and the metadata contained in the corresponding documents • Utilized the Whisper (Large) model to evaluate the performance of Columbia University’s Habanero High-Performance Computing (HPC) Cluster, assessing translation speed and accuracy across audio documents of 5, 20, and 50 minutes in duration.
• Architected a robust data pipeline using Python and MongoDB Atlas Data Lake, enabling seamless collection and cleansing of data for 91,000 documents; achieved fluency in MongoDB Atlas Data Lake within three days, elevating data management capabilities for the team • Analyzed and synthesized quantitative data using Tableau, resulting in conclusive reports that drove business improvements and cost reductions • Designed and implemented a classification model that enabled the product team to predict the necessary skills from job descriptions with a test accuracy of 93.8% • Communicated and collaborated closely with the 3 different teams to accomplish the Data Engineer project using required data collection and statistical models