Here are some of the things I've worked on!
At Demystifyd, our mission is to create a centralized platform for the international community of students and professionals to connect, progress professionally, and explore their careers. This summer I oversaw the launch of Demystifyd and its various features as a Product Manager. Demystifyd is now live from the link above!
I love sports and I love when Stanford is great at sports. Who makes these well oiled machines tick? Coaches. EzRecruit was developed as a Proactive Recruitment Management Platform designed to streamline and automate the recruitment processes for university varsity coaches. This project involved development of a web app that consolidates recruit information, automates data updates, and tracks interactions, significantly reducing the time coaches spend on administrative tasks. EzRecruit enhances the recruitment process, enabling coaches to focus more on coaching and building relationships with recruits.
Mixed Martial Arts (MMA) fans and bettors often speculate on the outcomes of fights, influenced by numerous factors such as fighting style, physical attributes, experience, etc. Accurate predictions can enhance decision-making and betting strategies. With 2 baseline predictors, a logistic regression model, and 2 neural network architectures. The logistic regression model achieved an accuracy of 66.4% and a precision of 71.4%. We also designed a simple feed-forward neural network (SimpleNN) and an improved version (ImprovedNN) with dropout layers for regularization.
Explored the application of the Quantum Approximate Optimization Algorithm (QAOA) to solve the Traveling Salesman Problem (TSP), a fundamental challenge in combinatorial optimization. Encoding the TSP into a quantum state using cost and mixer Hamiltonians and iteratively optimizing parameters to identify near-optimal solutions. Utilized the Cirq library for quantum circuit design and simulation. The project also deepened my understanding of classical and quantum algorithms, data visualization with Matplotlib, and graph theory. Implemented solutions for problem formulation where N=4, N=8, N=15.
This project leverages logistic regression to systematically evaluate and rank football players from the 2021-2022 season based on key performance indicators such as goals, assists, completed passes, and defensive actions. Utilizing Python’s pandas for data manipulation, numpy for numerical computations, and sklearn for the machine learning pipeline, the project preprocesses the data, including handling missing values, feature selection, and the creation of binary target variables. The logistic regression model, trained on a balanced dataset, demonstrated a 94% accuracy rate, achieving perfect precision in identifying top performers.
Download or view my resume below to get a comprehensive overview of my experience and skills.