# Yushu Zhou > Full-Stack Data Scientist Location: New York City Metropolitan Area, United States Profile: https://flows.cv/yushu A passionate, client-facing, and cross-industry data scientist. Able to translate ambiguous business problems to data science language and deliver the value of data science to business stakeholders. Love to take challenges. Projects range from credit reporting, HR practices to demand forecast, media buy. Programming: Python/ R/ Spark/ SQL Machine Learning: Predictive Modeling/ NLP/ Explainability/ AI Ethics/ Recommendation System Platforms: Watson Studio/ SageMaker/ Azure ML/ Databricks Know basics of kubernetes ## Work Experience ### Senior Software Engineer, agentic data applications @ LinkedIn Jan 2024 – Present | 纽约地区 ### Lead data engineer @ Thirdwave Jan 2023 – Jan 2024 | New York, United States - Managing data pipelines orchestrated by DBT that calculate wallet and dapp attributes from snowflake blockchain tables and copy them into AWS S3; - Managing cloud services including VPC, S3, lambda and alerting to load data into Redis - Developing our key Graphql subgraph in Typescript that exposes above attributes to REST API with <100ms response time (https://docs.thirdwavelabs.com/rest-api/request-parameters) ### Senior Data Scientist @ Coinbase Jan 2022 – Jan 2023 Coinbase wallet discovery & community ### Data Scientist @ Coinbase Jan 2020 – Jan 2022 | New York City Metropolitan Area Coinbase Earn -> Coinbase feed ### Advisory Data Scientist - Machine Learning, Data Science Elite Team @ IBM Jan 2019 – Jan 2020 | New York, New York Media Buy Clearance Prediction – Project Lead o Drove cross functional collaboration among engineers, UI specialists, and business stakeholders to improve clearance prediction model and enhance the analytical insights on client’s buy management platform for buyers; o Quickly iterated underlying prediction model and its deployment through python to Watson Machine Learning; o Demoed prototype of visuals on how buyers utilize model predictions to decide best unit price of bids and best daypart to place ads by Jupyter interact widgets; o Mentored junior data scientist on end-to-end model development pipeline; o Received satisfactory feedback from client. They'll continue and double software subscription soon. Continuous Monitoring of Productionized models o Designed architecture to embed fairness monitor, explanation provider and drift monitor of Watson OpenScale to evaluate live model performance from different perspectives; Implemented through python to Azure ML; o Successfully set up new monitors within client’s existing production pipeline for smooth operational change, resulting in 27 times software deal growth. Data Quality Auto-Detection o Identify key use case as contextual anomaly detection out of ambiguous descriptions from client; o Co-developed a method to detect contextual anomaly without domain knowledge infusion using PySpark on Databricks; o New method outperforms common algorithms on KDD Cup 99 data. Pending patent approval. Helped IBM develop strategic relationships with various clients o Built end-to-end data science workflows and presented them in meetings with clients to show value of IBM data science products o Joint data science discussions across different industries including retail, finance and media buy as data science expert. Provided vanilla solution prototype to our client during the meeting ### Data Scientist - Machine Learning, Data Science Elite Team @ IBM Jan 2018 – Jan 2019 | New York, New York Develop and deliver data science solutions to clients' business problems Auto Parts Demand Forecasting o Implemented a research method combining STL decomposition and LSTM using python (statsmodels and keras); o Stacked all models built by team and improved MAPE by 5%, compared to client demand forecasting software. Investment Manager Grading o Collaborated with subject matter experts to design key business KPIs for investment manager grading; o Applied KMeans clustering and shiny dashboard for better explaining investment manager ratings; o Provided counter-intuitive insights which were widely utilized into investment management workflow; o Successfully demonstrated values of data analytics to client, resulting in $1.8M deal. Aggregated Credit File Error Scanning o Developed binary classification system to replace rule-based file-error-check system using python to Watson Machine Learning. New system’s precision is around 40 times of baseline’s, which largely reduced the amount of manual work. o Designed a new method of resampling to mitigate extremely imbalanced data problem. ### Data Science and Web Scraping Intern @ Social Jan 2017 – Jan 2017 | Greater New York City Area www.socialtheapp.com 1) Routine Work Built automated web scraper to extract data from targeted websites in python. I learned a lot about HTTP request and response and got to know more about web application frameworks like AngularJS. 2) Movie Ratings Prediction Handled with over 500 mb data from Netflix. Wrote collaborative filtering from scratch to calculate similarity scores between any two movies. Because of the scale, deployed the program on a cluster in Linux. 3) Movie Content Features Test Built lasso model to test content features' explanatory power in python. Visualized results using matplotlib. ### Algorithm Engineer Summer Intern @ Alibaba Group Jan 2017 – Jan 2017 | Hang Zhou 1) Wholesale Goods Detector on B2C Marketplace Platform o Wrote regular expressions to pre-process over 101k item names (given by sellers) in Chinese for better text segmentation; o Built a Naive Bayes classifier in Python (pandas, numpy) from scratch to have more flexibility for model tuning, the vanilla version of which achieved nearly 93.54% F1 score; o Analysed misclassified cases and improved the model F1 score by 0.64%, mainly through fine-tuning parameters of posterior distribution and regularization; o The final model was highly recognized, integrated into one of the developing products. 2) Brand-Coupon Recommender System o Mined four types of user behaviors (view, favorite, add-to-cart, buy) on TMALL website (over 2TB data) and developed a two-level brand coupon recommender system: recommend personalized brands for users on the first level; matched coupons with users' per order value; o Tracked user behaviors in the following seven days to evaluate the recommender system: the accuracy of top 20, 10, 5, 2 recommendations were 52.83%, 40.74%, 28.97, 18.75%, respectively. 3) Routine work o Wrote SQL queries, web crawler and new UDF to extract required data. o Designed pivot tables or other visualization for data analysis to support business department. ### Research Assistant @ Investment & Finance Banking, Chinese Academy of Social Sciences Jan 2015 – Jan 2015 | Beijing City, China Researched about credit reporting system and compiled relevant articles about credit reporting system for our column "Financial Regulation Research International"; Interviewed some company owners to investigate real problems of building commercial credit reporting system in China and finally wrote an article to my mentor; Analyzed information regarding global and domestic economy for use in the China Development Bank Capital report. ### Wealth Management, 2015 Young Talent Program @ 花旗 Jan 2015 – Jan 2015 | Beijing City, China Learned about banking business and wealth management; Met with clients and helped them make investment decisions; Designed a new product and its marketing campaigns ### Financial Analyst @ AXA Jan 2014 – Jan 2014 | Hong Kong Based on the good team performances in three competitions - the improvement of tactical asset allocation, stock selection and hedge portfolio, our team won the chance of PTA in September. As the team leader during PTA, I successfully assigned task to every member and finished security analysis for our fund construction. I was responsible for data analysis (stock price and return rate prediction) during this internship for my team. ## Education ### Master of Arts (M.A.) in Quantitative Methods in Social Sciences - Data Science Focus Columbia University Jan 2016 – Jan 2017 ### Bachelor of management in Business Administration Tianjin University Jan 2014 – Jan 2016 ### Bachelor of Engineering (B.E.) in Resources Recycling Science & Engineering Nankai University Jan 2012 – Jan 2016 ## Contact & Social - LinkedIn: https://linkedin.com/in/jade-zhou - GitHub: https://GitHub.com/greenorange1994 --- Source: https://flows.cv/yushu JSON Resume: https://flows.cv/yushu/resume.json Last updated: 2026-03-22