Motivated Data Science undergraduate with a strong foundation in statistical modelling, data analysis, and visualisation.
As a Data Science undergraduate at UC Davis, I am passionate about building end-to-end machine learning models, exploring natural language processing, and turning complex datasets into actionable insights. My technical toolkit relies heavily on Python, R, and SQL, with hands-on experience spanning adversarial NLP pipelines, predictive logistics analytics, and optimized relational database queries.
Beyond technical development, I bring real-world corporate analytics experience from industry internships, along with proven leadership as a former Marketing Director and Project Lead. I thrive at the intersection of quantitative analysis and strategic problem-solving, where I can translate technical findings into high-impact business solutions and collaborate with cross-functional teams.
B.S. in Data Science
2023 – 2027
The Applied Financial Analytics Dashboard is an interactive Python-based platform engineered to execute automated quantitative evaluations and linear predictive modeling across fundamental enterprise data. Built using Python data engineering and visualization tools, the application translates multi-year financial disclosures into dynamic visual indicators and statistical metrics.
Centered on a preloaded machine learning analytics dataset spanning two decades of historical financial records (2006–2025) for major technology corporations like Apple Inc. and Alphabet Inc., the platform synthesizes data across income statements, balance sheets, and cash flow reports to evaluate key metrics such as revenue streams, net margins, return on assets, return on equity, and EBITDA.
To deliver forward-looking financial insights, the system incorporates an integrated Python-driven ordinary least squares (OLS) linear regression engine that calculates statistical parameters — including R² scores, t-statistics, p-values, and 3-year confidence interval forecasts — to quantify trend strength and evaluate projected growth trajectories.
In addition to pre-configured enterprise analytics, the platform features a flexible data ingestion module. Through an interactive upload workspace, users can import custom multi-statement CSV files for automated data parsing, feature extraction, and statistical regression calculation directly within the application runtime.
Engineered an end-to-end NLP credibility pipeline using a corpus of ~44K real-world news articles to evaluate news authenticity beyond simple binary labels.
Built and benchmarked a classical lexical model (TF-IDF + Logistic Regression) against a fine-tuned contextual Transformer (DistilBERT), achieving 99% test accuracy on held-out data.
Designed a probabilistic output framework featuring token attribution highlighting and calibrated confidence scoring to prioritize ethical, non-binary credibility assessments over false certainty.
Designed a predictive logistics pipeline in Python, with data seeding for edge cases.
Developed a localized shortage scorer and a demand trend analyzer.
Emitted JSON datasets for downstream dashboards.
I applied OpenAI frameworks to solve complex challenges by building and evaluating scalable, intelligent AI solutions.
I performed quantitative analysis on GRAMMY Awards media datasets using advanced Excel modeling to deliver strategic recommendations.
I designed interactive Tableau dashboards using Intel corporate datasets to transform multi-variable data into executive-ready visual insights.
I developed cross-cultural communication strategies and emotional intelligence frameworks to collaborate effectively across international teams.