Data Machine

Zukang Yang

********@*****.*** 949-***-**** www.linkedin.com/in/zukang-yang https://github.com/zukangy Skills

Languages: Python (Sklearn, PyTorch, Tensorflow, Keras, Flask, Zipline, Uplink), Git, SQL Database Management: MySQL, PostgreSQL, MongoDB (PyMongo/MongoEngine) Cloud: AWS S3, AWS EC2, AWS SageMaker

Machine Learning: Linear/ Logistic regression, Tree-based algorithms, Ensemble methods (XGBoost), K- Means, Dimensionality reduction techniques (PCA and SVD), ANN, CNN, RNN (LSTM) Statistics: Hypothesis Testing, ANOVA, Chi-Squared, T-test, Data Visualization, Data Wrangling, Feature engineering, Time Series analysis

Experience

Data Scientist - Eonum, Inc Irvine, CA Jan 2020 – Present ML-Driven Financial Trading System

● Develop a machine learning pipeline to predict stock market fluctuation leveraging market-specific data augmented with economic third-party sources (Quandl)

● Apply feature engineering to economic market data using techniques such as RSI, price momentum, and price acceleration to strengthen the model predictiveness

● Incorporate ML pipeline into the financial trading system to paper trade US equities for over 9 months, resulting in a 12% overall return and a 1.5 Sharpe ratio

● Leverage Python and SQL to optimize database workflow, resulting in a reduction of time required to update daily market data in the back-testing environment by 60% NER (Hydrogen Evaporation rate) Prediction

● Developed dynamic machine learning models in Python to estimate NER values for devices used to store human kidneys upon transplantation with data from device thermocouples

● Applied data processing techniques such as KNN imputation and power transformations, improving the model predictiveness from 70% to 93%, thereby streamlining the workflow of model development

● Deployed scalable and retrainable machine learning model to production using Flask on AWS EC2 to create intuitive user-friendly interface while abstracting ML complexity behind the scenes Data Scientist - Enhance IT Atlanta, GA Sept 2019 – Jan 2020 Single-family Home Appraisal Model

● Developed machine learning pipeline for a real-estate app ingestion to predict the fair value of single- family houses to reduce the expenses of hiring property appraisers and guide real estate investment

● Built a web crawler with Scrapy library in Python to scrape and process housing data off relevant websites; wrote queries to store the data in PostgreSQL database

● Model outperformed top competitor algorithm (Zillow) by leveraging an XGBoost model with hyperparameter tunings

Education

B.S. Mathematics, specializing in Data Science June 2019 Minor in Statistics

University of California, Irvine

AWS Certified Cloud Practitioner – In Progress

Contact this candidate