About Me

Hi, this is Giang (GiGi)!

I'm currently a Data Scientist @ Instapage. I am passionate about building impactful Machine Learning products and driving business strategies and growth through data-driven analysis. I have 3+ years of experience working in Data Science and Analytics at tech startups across multiple industries, including Education, E-commerce and Advertising.

I graduated with a Bachelor's of Science in Statistics at the University of California - Los Angeles (UCLA), where I have developed a strong programming , statistics and analytics skill set that can tackle business problems and generate meaningful insights for social impact.

As a Data Science intern (focused on Product Analytics) at Elsa Speak for approximately 2 years, my primary role is to work cross-functionally with product and marketing teams to connect mobile usage with business outcomes, manage A/B testing, answer ad hoc questions for product decisions, and conduct data analysis on user behavior to promote engagement and retention. .

In my portfolio below, I'd like to share some of my projects I completed so far that I found interesting. Hope you enjoy reading it!

Click "more" for details and source code on Github and Medium.

Featured Projects

Music & Movie Taste Analysis

An analysis and prediction on music and movie tastes bewteen couples based on Spotify and Netflix data


(Python, R, Seaborn, Random Forest, Logistic Regression, LDA)

MindfulR

Self-initiated NLP project on meditation app reviews, part of an on-going interest in democratizing mental health improvement – featured on Towards Data Science

(R, itunesR, tidytext, spaCy)

Company Data Analysis

Marketplace Analysis Case Study

Provided analysis on what types of pros customers are interested in and my recommendations for how the company can improve and grow their marketplace.


(Python, Seaborns, Logistic Regression, Random Forest, XGBoost)

Collaboration Platform Analysis Case Study

Focused on analyzing user adoption and key covariates for future adoption to provide recommendations for improving this experience and increase user engagement.


(Python, Seaborns, Logistic Regression, Random Forest, XGBoost)

A/B Testing

E-learning Platform - A/B Testing on User Conversion

Implemented an end-to-end A/B test analysis on how the changes in the asking users in advance if they have the time to invest in the course can help Udacity increase fraction of paying users after the free-trial.


(Python, scipy, statsmodels, seaborn)

Q&A Platform - A/B Testing on User Engagement


Initiated insights and recommendations for the the Q&A Platform to understand the impact of whether the new UI change would increase user engagement.


(Python, scipy, statsmodels, seaborn)

Data Hackathons

Expedia Customer Segmentation


Participated in the American Statistical Association (ASA) 2017, wrangled with a large, rich, and complex Expedia datasets to extract meaningful findings in on flights and hotel booking using K-means clustering and ultimately recommended targeted user segmentation for product and marketing stratergies within 48 hours.


(Tableau, Excel Solver)

Indeed - Tech Job Recommendation System

Analyzed and presented the Statistics capstone project on Indeed data to intiate recommendations for optimizing and addressing Tech job shortage problem to Statistic professor and 45 UCLA undergrads.





(R, data.table, plotly, ggplot, posterdown)

Facebook Data Challenge


Curated open-source data of San Francisco to make recommendations on opening new local businesses with a focus on social impact.






(R, ggplot, Tableau)

Macroeconomic Analysis

New Zealand - Social Network Analysis

Worked in a team of 5 aspiring Economic researchers and presented on the potential industries for New Zealand to enter using Social Network Analysis techniques and Harvard Atlas of Economic Complexity’s visualizations – received first place.


(Tableau, UCINET)

Norway - Time Series Analysis

Analyzed and presented the effect of exchange rate shock in Norway in a team of 4 Economics graduate students by using the structural vector autoregressive models in R



(R, ggplot, vars)

"All data has its beauty, but not everyone sees it."
-- Damian Mingle

Stay Connected