Utilized PyMuPDF, NLP, and byte-stream parsing to identify and extract objects on shipping container PDF documents.

Created a highly accurate time-series forecasting model with a MAPE of 3% using XGBoost decision tree modeling on a million row of data to be used for displaying current keyword search trends.

I conduct and perform analytical research on various aspects of university concerns.

Utilized Python, SQL, and Excel to develop automation pipelines to increase efficiency of budgeting tasks. Performed analysis using Python for next-year budget forecasting for resources in Google Cloud.

Developed data visualization and data cleaning pipelines using Python and SQL in Databricks, which incorporated data parsing and NLP techniques. Also created data insight tables using data inferences, sentiment analysis, and geospatial analysis.

Developed a sentiment analysis model on thousands of journal entries using Hugging Face to improve performance of a therapy journaling chatbot.

I led a small student team to build a foundation on creating a comprehensive database of the plant-based market. As part of my role, I facilitated communication between students and the Plant Futures organization, assigned roles to students, and also designed the project. As part of my technical role, I wrote various Python scripts to web-scrape and clean large amounts of financial information on over 200 plant-based companies which I then imported into and further cleaned in mySQL which is the foundation of the database that Plant Futures will be using and developing in the future. Following this, I used Apache Superset to visualize current financial trends from the data I collected occurring across various sectors in the plant-based market.