I'm currently studying Business Economics at UCLA. I'm interested in Math, Finance, and Computer Science.
June 2024 - August 2024
Worked as a paid intern under Dr. Mei-ching Fok at Goddard Space Flight Center. Analyzed the May 10th, 2024 storm. Described how far particle flux was able to penetrate L-shells. Compared Linear Regression and LSTM models' predictive power in describing expected proton and electron flux using solar wind parameters from satellite and ground observatories. Presented at the 2024 American Geophysical Society meeting
June 2023 - August 2023
Worked as an intern under Jun Dong, and Sirish Uprety. Used Tensorflow U-net, an image segmentation algorithm, to find the borders of iceberg A73. Processed data using PIL, Pandas, and Numpy. Training data was labeled by hand. Size, rotation, and location were calculated. Presented at the 2024 American Meteorological Society Meeting
July 2022 - November 2022
Worked as an intern under Dr. Surja Sharma. Created a Long-Short term memory(LSTM) model to predict Disturbance Storm Time(DST), a measure of the strength of Earth's Ring Current, using WIND satellite parameters.
Analyzed an open-ended business problem under time constraints and built a data-backed recommendation. Structured the approach using consulting frameworks, supported conclusions with quantitative analysis, and synthesized findings into an executive-ready deck delivered to judges.
View Project
Used Excel to showcase and compare cost allocation methods. Discussed their relation to corporate governance. Implemented a novel "Lattice Allocation" method (Bent, Caplan 2017) based on matrix principles of Linear Algebra. Results were within 0.1% error after only 3 operations.
View Project
Designed a lightweight mathematical computing library for symbolic-style function representation, numerical evaluation, and statistical analysis. Implemented a tree-based expression engine supporting differentiation, Taylor approximations, and probability utilities to streamline quantitative experimentation and modeling. Implemented probability utilities (PDFs, CDFs, inverse-CDF tables) using numerical integration and error-controlled approximations.
View Project
Developed an experimental neural-network framework from scratch with explicit backpropagation and graph-based visualization. Combined physics-inspired graph dynamics with gradient and stochastic optimization methods. Built multiple training strategies including gradient descent, Metropolis-Hastings random walk, and Monte-Carlo–style sampling.
View Project
Designed & implemented a recursive algorithm–based simulator to generate fractals and compute Hausdorff dimensions via box-counting. Applied fractal analysis to natural features (e.g. coastlines and borders) for quantitative modeling and visualization. Achieved <0.6% error on known fractals.
View Project