GradQuest helps students save time and reduce stress by matching them with graduate programs that fit their personal constraints such as budget, location, and GPA, GRE etc..The platform streamlines the entire application process by  tracking applications and their requirements, organizing documents, and sending timely reminders for upcoming application deadlines. 

• OAuth2 login (Discord / GitHub) for secure sign-in

• React Native, TypeScript, JavaScript, Springboot

• Student user profiles with preference intake (target countries, budget, funding needs, deadlines, etc.)

• Generates a top-5 program match list based on user preferences (drawn from 60+ universities in the database)

• Saved Applications page to keep track of the schools you plan to apply to

• Reminders page to view and receive notifications for important application deadlines and requirements

​

We wanted to make our in-house API instead of using an online one that is premade.  Using a premade front-end, we configured our own API for a vocab app with Google OAuth.

• Developed a cross-platform vocabulary app using React Native, enabling users to create, save, and manage vocabulary lists across Android, iOS, and web, improving accessibility for all supported platforms.

• Designed and built a custom REST API with Spring Boot and MongoDB, supporting CRUD operations for users and vocabulary lists.

• Implemented Google OAuth authentication using expo-auth-session, enabling secure user login across devices and improving authentication reliability by 75% compared to manual login.

• Optimized data handling with AsyncStorage caching and automated testing via Jest + jest-expo, achieving 85% test coverage.

• Collaborated via GitHub and Slack to deploy the backend API on Heroku.

​​​

• Analyzed New Orleans Police Department’s Use of Force dataset to predict force levels using officer, subject, and contextual features (e.g., race, service years, officer count, proximity).

• Preprocessed data by handling missing values, encoding categorical variables, and engineering features; applied and compared ML models including Logistic Regression, Decision Trees, and K-Nearest Neighbors (KNN), with KNN achieving 67.8% cross-validation accuracy.

• Identified race, ethnicity, and number of officers present as key predictive factors; evaluated class imbalance and contextual complexity to assess model limitations and inform real-world applications.

This project involves designing a MySQL database management system for a fast-food franchise to efficiently track food orders, inventory, supplies, rewards, transactions, and more across international locations.

• Technologies Used: MySQL for creating and executing queries.

• Key Deliverable: Developed a comprehensive ER Diagram to model the database structure.

Want to put filters over your pictures just like you do in Apple Photos? Well me and my team created a web platform to fetch random images from external APIs and apply diverse image filters utilizing Pillow and Python.

• Python

• Flask

• Bootstrap

• Pillow

As seniors who are actively look for internships and full time jobs, my team and I wanted to make a tool that can help us find job applications all in one place using the Adzuna API. 

• Developed a cross-platform job listing app using React Native that enables users to register, log in, edit profiles, and manage saved jobs, improving mobile job-search efficiency by 93% through streamlined workflows.

• Integrated the Adzuna API to deliver real-time listings from over 1 million job postings, expanding data coverage and accuracy for users.

• Implemented persistent local storage with AsyncStorage, reducing login time by 85% and enabling offline access to saved jobs.

• Wrote 25+ Jest unit tests, achieving 90% test coverage and reducing post-merge bugs by 74%.

• Collaborated with a 4-person agile team using GitHub pull requests and Slack, completing 12+ user stories and delivering all core features two weeks ahead of schedule.

​​​

With countless animals and limited management tools available, veterinarians often struggle to keep track of critical details. To address this challenge, my team and I

• Designed and developed a full-stack web application for managing a veterinary clinic’s animal visits and health records.

• Implemented secure user authentication using bcrypt and sessions using Express and developed a robust backend with Node.js and MySQL to support full CRUD operations.

• Oversaw project planning, delegated tasks, set deadlines, and ensured team productivity and collaboration throughout development.

​

​

​Using pandas, matplotlib, and seaborn my team and I 

• Conducted exploratory analysis on Sonoma County Animal Shelter data to examine how breed, intake type, and other factors influence adoption rates and shelter stay duration.

• Cleaned and preprocessed the dataset by removing incomplete records and converting columns to appropriate data types; used groupby and crosstab to aggregate and compare adoption trends.

• Visualized key insights using bar plots and histograms to highlight breed-specific stay lengths, intake effects, and outcome patterns, helping inform policy and outreach strategies.

This is an android app that manages the operations of a library. This includes creating an account, doing book reservations, and system management for admins. The technologies I used were :

• Java

• Android Studio

• SQLite