Gregory Eales Email & Phone Number

San Diego, Ca, Us

San Diego, California, United States

- Redesigned and implemented an ML-based fraud detection system, reducing fraudulent withdrawals by 75% and reducing overall losses by $1.3 million annually. Employed Python, scikit-learn, Pyplot, and Pandas for data analysis, cleaning, and feature engineering. Trained and validated a Gradient Boosting model and then deployed with multi-threading on a Linux (CentOS) server with a RESTful Flask API Service, achieving a response time 10x faster compared to the previous legacy system.- Performed Session and User level A/B testing for changes to UI and backend functionality as well as new ML model roll outs. Applying statistical tests to collected data and reported on engagement, retention, and revenue to upper management.- Successfully implemented the data collection phase of a PyTorch-based ML pricing recommendation system, utilizing JavaScript and AJAX for recording user price offerings and actions into a MySQL database. Developed a training pipeline using PyTorch to model user behavior, contributing to deeper user engagement insights. Currently focusing on completing the dynamic pricing model and preparing for deployment stage, with the aim of optimizing user incentives for enhanced revenue growth.- Spearheaded the end-to-end creation and execution of a machine learning pipeline, including performing critical ML tasks such as feature selection, model training, hyper-parameter tuning, and deployment. This resulted in a highly accurate user churn prediction model with a 82% accuracy rate, markedly enhancing data-driven strategies and customer retention.

San Diego, California, United States

San Diego, California, United States

San Diego, California, United States

Carmel, California, United States

Bachelor'S Degree, Computer Science

High School Diploma

Activities and Societies: AP Scholar Award