Brian Xu D.

About Brian Xu D.

I worked with large graph data on distributed systems for information diffusion research during my doctoral studies. Along with graph analytics, I have internship experiences in using machine learning techniques to analyze time series (panel) and cross-sectional data. I build ELT/ELT pipelines using Spark and Airflow to productionize my ML models. I used Tensorflow to build Deep Neural Networks to interpret images or natural language data. Here are the tools/skillsets I frequently use for these projects- Data Science: Statistical Analysis (Time Series, Hypothesis Testing, Bayesian), Machine Learning (Linear and Logistic Regression, Random Forest, Clustering, Regularization, PCA, Feature Engineering, Dimensionality Reduction), Deep Learning (CNN, RNN, LSTM, GAN)- Programming: Python (pandas, scikit-learn, Jupyter, NLTK, genism, TensorFlow, Keras, PyTorch, PySpark, XGBoost, Prophet, NetworkX, Dask), R (dplyr, data.table, igraph), Scala- Database/ETL: SQL, NoSQL (Cassandra), Data Warehouse (Redshift), Data Lake (EMR), Data Pipeline (Airflow)- Distributed Systems: Spark (DAG, RDD, DataFrame and GraphFrame APIs, spark.sql, spark.ml), Hadoop (MapReduce, HDFS), AWS (EC2, S3, Elastic Beanstalk)- Miscellaneous: Web Scraping (BeautifulSoup, HTML), Visualization (seaborn, matplotlib, ggplot2, dash, plotly), Version Control (git), Product Deployment (Flask, Heroku)Check out the data product I built during my recent fellowship program: A web application that use machine learning to help online Peer-To-Peer (P2P) lenders predict and track their ROI on lending investments.http://aproduce-env.y8hx22nuj3.us-west-2.elasticbeanstalk.com/

Experience

1. Data Solution Architect

   Starburst

   Dec 2020 — Present · Seattle, WA, US

Education

• University of Miami

  Doctor of Philosophy (Ph.D.), Industrial Engineering

  2013 — 2019