About

I am a fourth-year student at the University of California, Berkeley, pursuing a double major in Data Science and Applied Mathematics, with a concentration in Numerical Analysis. Currently, I work at Omnium, a data science consulting firm in the consumer packaged goods industry, where I analyze trends in data and provide insights that support client decision-making. Looking ahead, I am primarily interested in contributing to the data science, machine learning, and artificial intelligence domains, applying my mathematics background to enhance my work in these areas. Feel free to reach out if you'd like to discuss my experiences or the worlds of data science and math in general!

Education

University of California, Berkeley • Fall 2022 - Spring 2026 • GPA: 3.85

Data Science (BA) - Applied Math and Modeling concentration

Applied Mathematics (BA) - Numerical Analysis concentration

Experience

Data Scientist Intern → Data Scientist (Part-Time)

During my summer internship at Omnium, I contributed to data-driven consulting projects in the consumer packaged goods industry, bridging client data with real-world business strategy through mathematical, data science, and business methods. I designed a research project and Python pipeline that introduced seasonality-based clustering algorithms to strengthen pricing regressions and improve demand modeling. I also built and compared regression models to quantify market preferences, projecting expected sales growth from new product launches and creating a scalable framework for analyzing brand innovation. Looking ahead, I am continuing with Omnium as a part-time data scientist, where my focus is on automating and optimizing the pipelines that transform raw data directly into client-facing metrics, visuals, and tables.

Data Intern

At the Fung Institute, I work to uncover insights from student and alumni data, driving improvements and highlighting the impact of initiatives such as the Fung Fellowship and Master of Engineering programs. Using tools like Salesforce and Google Workspace, I manage and refine databases to ensure accurate and efficient data tracking. My role also involves creating compelling data visualizations that effectively communicate findings, enabling collaboration with teaching teams and stakeholders to advance the institute's mission and share the value of its programs.

Data Analyst

As a Data Analyst, I leveraged SQL, Python, and Tableau to query, clean, and visualize student demographic and academic data for strategic enrollment decisions. Collaborating with department heads and fellow analysts, I regularly translated complex data into actionable insights; for instance, in 2024, my analysis contributed to enrolling over 800 additional in-state students (a 1% increase from previous years). I also developd interactive, data-driven visualizations for EM's website, writing accessible HTML code to enhance user experience and highlight key trends for stakeholders.

Admissions Exam Reviewer

At SUMaC, I evaluated admissions exams for Stanford’s most competitive high school math programs, assessing students’ mathematical creativity, logical reasoning, and proof-based problem solving skills, beyond rigid rubrics. My role involved providing comprehensive written evaluations that highlight each applicant’s strengths, weaknesses, and problem solving approach, directly shaping admission decisions for top candidates.

Research

Seasonality Clustering — A Hierarchical Agglomerative Approach

As part of my internship at Omnium, I designed a research project exploring seasonality-based clustering methods for pricing analytics. The question I proposed was: "How can we replace fixed quad-period seasonality with data-driven seasonal clusters to reduce pricing regression complexity and capture true demand variation." To answer this, I implemented two hierarchical agglomerative clustering pipelines - one that clustered weeks just based off of their average dollar velocities, and one that also included distance between weeks in a custom distance metric in order to prioritize continuous clusters. I evaluated these models across seven linkage criteria (focusing on single linkage), using dendrograms, silhouette scores, Calinski–Harabasz indices, and Davies–Bouldin indicies to quantify results. Ultimately, I applied these clustered season definitions to pricing regressions, resulting in reduced multicollinearity compared to standard 13-period models and improved demand predictions. View the summary slides or explore the code.

Analyzing Academic Resource & Funding Allocation at UC Berkeley

Under the supervision of Dr. Stefano Bertozzi, I conducted an analysis of UC Berkeley's financial landscape for the Academic Senate's Committee on Academic Planning and Funding Allocation (CAPRA). I queried and analyzed campus-wide financial and enrollment data from UC Berkeley's central database, CalAnswers, utilizing Excel to visualize and interpret revenue streams and expenditure trends across undergraduate and graduate programs. My findings informed CAPRA's strategic planning and resource allocation decisions.

Exploring Indian Non-Governmental Organization (NGO) Distribution

Through a collaboration with Daanmatch, I led a project that identified geographic areas with high NGO concentrations, guiding the strategic allocation of funding and support. I performed exploratory data analysis in Python (Pandas, Seaborn, Matplotlib), constructed data pipelines for cleaning/analysis, and standardized and mapped the addresses of over 10,000 NGOs across India, primarily using ReGex for address formatting. I presented the team's findings in a poster at the Data Science Discovery Program Symposium. View the project report here.

Projects

Omnium - Flavor Analysis

Designed a pipeline to estimate new flavor performance by creating velocity-based indices across four methods, including category averages, percentile rankings, and a brand-controlled regression model. Tested multiple regression configurations—varying intercepts and log transformations—to produce interpretable, scale-free flavor indices. Final model was implemented in Python in Excel for flexible filtering and output. View the summary slides and a detailed project brief.

Stack Overflow Survey Dashboard

Built an interactive dashboard in Sketchingpy to visualize how education levels influence salaries and technology usage across industries, using self-cleaned and processed 2024 Stack Overflow Developer Survey data. Applied object oriented design principles in the dashboard code to structure graphs, interactivity, and user inputs clearly and modularly. The project includes preprocessing scripts, datasets, and the full Sketchingpy project file for reproducibility. View all project materials on GitHub.

Spam Email Classification

Built a logistic regression model to classify more than 8,000 emails as spam or not spam, achieving 85% test set accuracy through feature engineering and model tuning. Leveraged RegEx to parse email text, engineering features like custom word presence indicators and punctuation frequency. Evaluated performance with precision, recall, and ROC curves, iteratively refining the feature set and optimizing hyperparameters to minimize false positives and enhance overall reliability.

Cook County Housing Price Prediction

Developed a predictive model for housing prices using linear regression and a custom data pipeline on 500,000+ records. Performed extensive feature engineering—including outlier removal, log transformations, and one-hot encoding—while analyzing potential biases to ensure fair and accurate results. Validated the model’s performance on real-world data, demonstrating the effectiveness of exploratory analysis and iterative refinement in delivering actionable insights.

Bunny Trails Game

Built a 2D Java game (2,500+ lines of code) featuring randomly generated worlds driven by user-provided seed numbers. Implemented multiplayer modes and interactive mechanics, incorporating mouse and keyboard inputs for character control. Integrated PNG/GIF for graphics, OTF for fonts, and WAV for music, and stored player progress locally via a text file to maintain persistent gameplay.
Downloads for Mac and select Windows versions can be found here.

Gitlet

Recreated core Git version control functionalities—including add, commit, branch, and merge— entirely from scratch in Java. Designed and implemented internal data structures to track file states, handle branching logic, and manage snapshot history, while ensuring error handling and workflow consistency. Focused extensively on system architecture, object-oriented design, and testing to deliver a fully functioning mini Git system from the ground up.

Ants vs. SomeBees

Implemented a multi-phase Python project emphasizing object-oriented programming, inheritance, and composition through replicating the video game "Plants vs. Zombies". Developed specialized classes of 'Ants' and 'Bees' that override and extend base classes, manage dynamic game states, and interact seamlessly with each other. Strengthened debugging and testing skills by writing local tests and iteratively refining the game mechanics.

Modeling π and e through Approximation and Application

Collaborated on a project to approximate π and e using both numerical and geometric methods in Python. Leveraged these approximations in a damped pendulum physics simulation to demonstrate real-world applications of fundamental constants, highlighting the practical intersection between theoretical math and applied science. View the project code, report, and presentation summary here.

Extracurriculars

Student Advisory Board Member

As a board member, I advise the College of CDSS on student needs and emerging concerns by participating in regular pulse surveys and providing feedback on core initiatives. I contribute to discussions around advising processes, diversity and inclusion, and student organization support, ensuring that student perspectives shape college-wide decisions.

MPS Scholars Program Mentor

In this role, I mentor undergraduates majoring in math, holding biweekly group sessions and one-on-one meetings to offer tailored academic advice and share my personal experiences. My goal is to foster a supportive community within the MPS department, helping students plan their academic/professional careers and stay connected to valuable resources.

Skills

Coding Languages: Python, Java, SQL, R, HTML, RegEx, LaTeX

Libraries: Numpy, Pandas, Scikit-learn, Scipy, Matplotlib/Seaborn, Sketchingpy

Data Analysis Tools: Excel, Tableau, Google Sheets, ATLAS.ti

Specializations: Python in Excel, Clustering Algorithms, Data Pipeline Optimization

Languages: Spanish (California State Seal of Biliteracy)

Relevant Coursework

Data Science: Foundations of Data Science, Computational Structures in Data Science, Principles & Techniques of Data Science, Data Structures and Programming Methodology, Numerical Analysis for Data Science, Probability for Data Science, Data Engineering, Data Mining & Analytics

Mathematics: Multivariable Calculus, Linear Algebra & Differential Equations, Abstract Algebra, Real Analysis, Complex Analysis, Numerical Analysis