Projects
Chrome Extension: Better BrightSpace
Overview
Research and Development phase.
- Utilized Brightspace APID2L Authenticated API for developers.
Algorithmic Analysis of Antimicrobial Prescription Patterns
Overview
This project investigates antimicrobialAgents that kill or inhibit the growth of microorganisms, including bacteria, fungi, and viruses. prescription patterns in adults and children, analyzing over 50,000 prescription records to identify trends and improve stewardship practices.
- Utilized Principal Component Analysis (PCA)A technique to reduce data dimensionality while preserving variance..
- Applied clusteringGrouping data points into clusters based on similarity..
- Analyzed data from 2021-2022.
ReUseVandy: Student Marketplace Web App
Overview
This project implements a three-tier architectureA design pattern that separates the application into Presentation, Logic, and Data layers for modularity and scalability. for a student-exclusive marketplace to resell goodsSecond-hand items like textbooks, electronics, and furniture, tailored to college students' needs.. ReUseVandy enhances the campus community by promoting sustainability and convenience.
- Built the front-end using ReactA JavaScript library for building fast and dynamic user interfaces. to provide a smooth user experience.
- Implemented a secure and scalable back-end with Node.jsA runtime environment for executing server-side JavaScript code..
- Leveraged PocketBaseAn open-source backend as a service (BaaS) for lightweight database management and authentication. for real-time data synchronization and user authentication.
- Deployed the app using NetlifyNetlify.com for hosting and managing scalable cloud resources. for reliable performance.
Arduino-Driven API for Stock Option Strategies
Overview
This project integrates an ArduinoA microcontroller platform that enables hardware-software integration for automation and data transmission. with a custom API to deliver optimal stock option strategiesStrategies involving financial instruments based on stock price predictions, such as calls and puts.. Built upon concepts from the CaltechX BEM 1105x course, the system evaluates real-time conditions to suggest strategies at the best-performing intervals.
- Developed a custom API to transmit real-time stock predictions from RobinhoodA popular platform offering financial data and trading capabilities. and TradierAn API providing comprehensive stock and options data..
- Implemented machine learning modelsStatistical models such as Brownian motion simulations and Black-Scholes for options pricing..
- Utilized financial APIs to fetch real-time market data and integrated these predictions into the Arduino system.
- Optimized strategy intervals using queueing systemsMechanisms to manage and prioritize API requests, ensuring timely updates..
Chocolate Consumption and Nobel Laureates
Overview
This study explores the intriguing hypothesis that higher chocolate consumptionMeasured in kilograms per capita, representing a country's average chocolate intake. is associated with increased intellectual output, as proxied by the number of Nobel laureates produced per 10 million people. Latitude is used as an instrumental variable to investigate this relationship.
- Utilized Two-Stage Least Squares (2SLS)An econometric method used to address potential endogeneity by using instrumental variables. to analyze the causal relationship.
- Data included variables such as latitudeAbsolute distance from the equator, influencing cocoa cultivation feasibility., research expenditure, and scientific output.
- Built models to predict chocolate consumption using latitude and estimated its impact on Nobel laureate production.
- Controlled for confounding factors such as GDP, research funding, and scientific publications to isolate the effect of chocolate consumption.
Strategic Analysis of Climate Change Negotiations
Overview
This project explores the dynamics of climate change negotiations through the lens of game theoryA field of mathematics that models strategic interactions among rational decision-makers.. Using concepts such as sequential and simultaneous games, mixed strategies, and repeated games, we analyzed the interplay between nations in addressing climate change challenges.
- Developed a two-player gameA simplified representation of interactions between nations with strategies to cooperate or defect. with assigned payoffs reflecting economic and environmental impacts.
- Incorporated subgame perfect equilibriumA refinement of Nash equilibrium ensuring rational decision-making at every point in a game. to analyze multi-stage decision-making scenarios.
- Used quantitative metrics such as payoff matrices and rollback equilibrium to model strategic decisions in defection and cooperation scenarios.
- Integrated real-world variables, including climate financing and renewable energy investments, to contextualize the theoretical framework.