9+ years of experience in applied statistics and machine learning, but my specialties are in time series and user-behavior data.
I currently use Python, SQL, bash, and sometimes JavaScript.
Whether through words or pictures, I’ve spent years crafting my skills in communicating to diverse audiences. Check out my writings in my Blog!
While I intrinsically enjoy solving problems, I am driven most when my work has positive impact.
BSc in Bioinformatics and Psychology
MA in Psychology
MSc in Computational Cognitive Neuroscience
Under construction—please stay tuned for updates as I put the finishing touches on the work that I’d like to share!
The dashboard presents visualizations and analysis of Fitbit fitness and health tracking data collected from my Fitbit Charge 4 from November 1, 2021 to March 3, 2022. While I started creating it because I wanted to know whether there were any relationships between fitness and health metrics, it later became an exercise in creating a… Read more…
Does anyone remember the fall of 2017? That’s when HQ Trivia took the world by storm. This live mobile trivia game initially featured 10 questions and ran only twice a day, attracting millions of eager players vying for cash prizes. My friends and I were no exception, gathering together in a Discord channel every day… Read more…
A niche free-to-play online survival game was facing the problem of diminishing annual revenue over time and was seeking new avenues for monetization and retention. Leveraging a dataset of in-game user activities, load-outs, and states, I employed clustering algorithms to segment players into distinct groups based on their behavior, preferences, and engagement. The analysis… Read… Read more…
DonorChoose.org is a non-profit crowdfund platform for teachers to request resources for their classrooms. Founded in 2000 by a public school teacher and largely operated by volunteers, the website faced a scaling issue as the number of proposals grew exponentially every year. The solution was to use an algorithm to pre-screen applications and provide a… Read more…
Deep learning is powerful but often a black box. To understand predictions, we need insight into what our models have learned. In this project, I examined the patterns of distributed activity in a simple Recurrent Neural Network (RNN) with a context memory layer. To simulate a toy Natural Language Processing (NLP) task, I trained my… Read more…
Modeling changes in intraday trading volume is a key component in algorithmic trading strategies. Deep learning can be a useful tool for discovering new edges for investors, but for several reasons the use of recurrent neural network models in financial data have produced mixed results. In this on-going project, I explore the application of deep… Read more…
Winner of the “Most Beautiful” category of Becoming a Data Scientist’s first-ever #CraftyDataViz contest!
I recreated Circos diagrams using needle and thread. Circos plots are useful for visualizing relationships in a circular format. (Original report here)