I am currently looking for data science roles, either remote or in the Minneapolis area. I am passionate about the life sciences, improving healthcare outcomes in the United States, and building data processing tools for scientists.
I am a neuroscientist and engineer with over 9 years of experience in data analysis, machine learning, and computational modeling, resulting from my work at the University of Minnesota and UMKC. My PhD research project demonstrates excellent project management skills in which I planned and executed short and long-term goals and consistently achieve key deadlines. This has helped me become a self-motivated worker who enjoys problem solving and seeking out learning opportunities. I am comfortable in a collaborative environment, where my teamwork skills and technical expertise helped me thrive in several cross-functional teams which included physicists, biologists, and engineers.
Programming: Python (6+ years), SQL, MATLAB, R, C/C++, Git
OS: LINUX, Windows OS
Technical Skills: Biomedical Data Analysis, Neural Engineering, Signal Processing.
High Performance Computing (University resources, CMRR and Minnesota Supercomputer Institute).
Machine Learning, Algorithm Development, Large Dataset Management, PyTorch.
Statistical Modeling, Un/Supervised Learning, Clustering, GLM, Regression, Network Analysis.
My thesis work explored the network structure of spontaneous activity in the developing visual cortex. This required collecting a big dataset of calcium imaging movies and setting up a dataframe to manage it. Over the course of the project, I wrote a custom Python library which preprocessed this dataframe of movies, computed key metrics, generated synthetic datasets, and performed statistical tests. Using unsupervised clustering and other machine learning methods, we found that spatiotemporal patterns in spontaneous activity repeated across hours and predicted future activity.
Dataset management of novel spontaneous activity dataset, where class functionality mimics NumPy, but for arbitrarily long temporal sequences.
https://github.com/LunaKet/SequenceDataClass
Analyzing spatiotemporal pattern predictability over time. Demonstrates deep knowledge of parallel processing on high performance computing clusters.
https://github.com/LunaKet/SpatiotemporalClassiferHighCompute
Projects completed at the University of Missouri-Kansas City. While beginner-level, these projects show both my growth from a nascent programmer, as well as demonstrating a wide range of basic proficiency in C++, PyTorch, opencv, and firmware development.
VNC analyzer for biometric research project (firmware):
https://github.com/LunaKet/MiniCircuits-2x8-Switch-Automation
PyTorch project for Computer Vision course:
https://github.com/LunaKet/pytorch_tripletloss
Facial recognition project for raspberry pi + arduino setup:
https://github.com/LunaKet/NerfGunFaceRecognition