Skip to content
View lzhao-byte's full-sized avatar
2 followers · 7 following
  • Portland, OR

Block or report lzhao-byte

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lzhao-byte/README.md

Hi, I'm Liuhui Zhao 👋

Data analytics lead in Portland, OR, with 10+ years of experience with SQL/Python pipelines, data modeling, and analytics products. Today I work mostly in Snowflake: layered raw → QA → production pipelines, stored procedures / tasks / UDFs, semantic views, and self-service Streamlit apps. Earlier I built Hadoop/Spark and geospatial ETL pipelines for transportation research. PhD, M.S. Analytics (Georgia Tech). SnowPro Core · AWS Certified Data Engineer – Associate.

Most of my professional work lives in private company repos. The repos below are personal and teaching projects.

Featured

  • Intro to Predictive Analytics: a multi-page Streamlit app I built to teach the predictive-analytics lifecycle in company-wide data-literacy training. It walks through data exploration, cleaning, feature engineering, model training (linear models, Random Forest, XGBoost), and a simulation of model monitoring and data drift. Built with Polars, scikit-learn, Plotly, and an optional Snowflake backend. Live demo
  • Denver Housing Analytics: about 200K property sales joined spatially with neighborhood features (crime, crashes, tree cover, distance to amenities), plus price models (Lasso, Random Forest, XGBoost) and a Plotly Dash web app. Georgia Tech team project.
  • Research Topic Modeling: scraped 2010–2020 transportation journal papers, then applied LDA topic modeling (Gensim) and co-author network analysis (NetworkX).
  • workcal: company holiday policies as TOML config, turned into workday counts, WORKDAY()-style date math, and a dim_date table. It ships as a Python library, a CLI, and a Streamlit app, with 103 tests: regression tests against v1 and cross-checks against official US/UK calendars over 41 years. CI runs on GitHub Actions. Live app

Tools: SQL · Python (Polars, pandas, scikit-learn, XGBoost) · Snowflake (Snowpark, Tasks, Streamlit in Snowflake, Cortex) · Spark/Hadoop · AWS · Streamlit · Plotly/Dash · Tableau · ArcGIS / GeoPandas · Git + GitHub Actions

LinkedIn · Google Scholar · Website

Pinned Loading

  1. workcal workcal Public

    Company holiday policies as TOML config → workday counts, WORKDAY()-style date math, and a dim_date table. Python library, CLI, and Streamlit app.

    Python

  2. denver-housing denver-housing Public

    Denver housing analytics: 62K sales joined spatially with neighborhood features, XGBoost price model (R² 0.89), Plotly Dash app with MapLibre maps. Tested, CI on GitHub Actions.

    Jupyter Notebook 1

  3. topic_modeling topic_modeling Public

    LDA topic modeling + co-author network analysis of 2010–2020 transportation research papers (JITS, TITS, TRC)

    Jupyter Notebook

  4. used_car_price_predictions used_car_price_predictions Public

    Interactive Streamlit app teaching the predictive-analytics lifecycle: explore → clean → feature engineering → model training → drift monitoring

    Python