About Me
I was born and raised in New Orleans, where I attended Tulane University, receiving my BS and MS degrees in Neuroscience. In 2010, I began creating my career as a scientist and programmer by building my first data processing architectures to support analysis of complex neurobehavioral datasets.
In 2015, I moved to New York and joined a lab with far more data: tens of thousands of participants with neurobehavioral, genealogical, and genetic information. I co-engineered a database system to store, update, and access 100k+ records from disparate sources, and I wrote a module for analyzing and plotting EEG data. I utilized the tools I made to create visualizations supporting the lab's grants and publications.
From 2017 to 2022 I completed a Ph.D at UCLA, where I focused on creating computational models that could explain neural functions. Through my work, I gained foundational knowledge of statistics and machine learning from the neuroscientific perspective, and I continued refining my skills in handling large and complex datasets.
In October of 2022, I accepted a position as a Data Scientist in the Precision Agriculture division at FMC. There, I help create predictive models that enable farmers to anticipate the pressure of pests, weeds, and diseases on their crops. This allows farmers to minimize the usage of pesticides, herbicides, and fungicides, which minimizes costs, environmental impacts, and the development of biological resistances.
Skills
Projects
Database Detective: Discogs
I explain how I turned the largest online music database, Discogs, into a local SQL database on my machine. Using it, I answer some interesting questions, make some fun top 10 lists, provide a brief history of recorded audio, and conduct a network analysis of musical styles.
Is Dale's Law Computationally Beneficial?
In artificial neural networks, a single unit can excite some of its targets and inhibit others. But in biology, neurons are almost always exclusively excitatory or inhibitory (Dale’s Law). Do networks of excitatory and inhibitory units provide a computational benefit? In this research project I use PyTorch to show that when given the opportunity, recurrent neural networks (RNNs) will train themselves to follow Dale’s Law to accomplish certain simple tasks.
HBNL Database
While working at a lab in Brooklyn, I co-wrote a Python module for building, updating, and accessing a lab database implemented in MongoDB. Still used today, it contains hundreds of thousands of records from disparate sources (EEG, behavioral experiment, survey, genetics, genealogy).
AirBnBs in New Orleans: Tableau Study
Using a public database of AirBnB listings, I created this interactive visualization in Tableau. It shows AirBnB listings in my home town of New Orleans. Point size is based on the number of reviews, and point color is based on the price per bedroom. The user can filter listings shown based on the number of beds, bedrooms, and the room type. Hovering a listing shows key info in a tooltip.
Scientific Visualizations
Over my eleven-year career as a scientist, I've created hundreds of data visualizations for manuscripts, presentations, and grants in order to support lab goals. Here, I show a gallery of some of my favorites. The visualizations were generated using Python, R, and MATLAB, and editing was done in Adobe Illustrator.
Exquisite Text
Exquisite Corpse is a game invented by surrealist artists in the 1920's. Each player adds to a composition in sequence, by being allowed to see only the end of what the previous person contributed. Here, I implemented an online multiplayer version for poetry.
