Work Experience


American Express

Software Engineer
February 2023 - Present
  • Collins accepted a full-time position at American Express, and is currently part of the Digital Acquisitions organization within AMEX

American Express

Software Engineer Intern
June 2022 - August 2022
  • Uplift application portal for international customer onboarding using ReactJS and Java backend.
  • Architected and built custom React hooks for passing critical rendering information from one component to another.
  • Built REST API with CRUD operations in Java for parsing payloads from API calls and writing information to PostgreSQL database
  • Wrote a query builder class using builder patterns in Java to streamline and unify PostgreSQL query execution
  • Frequently directed team of eight co-interns on what tasks to do and how to execute them to advance project requirements and meet sprint deadlines
  • Contributed documentation for setting up SSH keys and Git/GitHub to the company-wide developer handbook
  • Wrote documentation on setting up various application backend functions, including the REST AP
  • Founded Intern Spotlight, and program at AMEX that seeks to highlight the background of various interns in the company

Data Science for Social Good

Research Fellow
June 2021 - September 2021

The Data Science for Social Good UK program (DSSGx) is a 12-weeks internship experience where research fellows solve challenging problems facing humanity today, leaving a lasting impact on the world. I was a research fellow this past summer, where I worked on a project aimed at prioritizing environmental complaints received by the environmental agency of the government of Chile. The goal was to classify complaints with the highest environmental impact so that the agency can send analysts to the site for inspections and fining whoever was at fault, thereby reducing environmental impact. I worked on a team of four persons, and my responsibilities are summarised below:

  • I led the data engineering aspects on a team of four to build a prioritization model for classifying environmental complaints using Scikit Learn, achieving an accuracy of 90% and an estimated 80% reduction in complaint analysis time.
  • Wrote a data pipeline for preparing classification data from CSV files to optimized numpy matrices.
  • Deployed model on Azure Cloud with detailed documentation hosted on GitHub Pages using mkdocs.

Center for Security and Emerging Technology

Data Research Analyst

The Center for Security and Emerging Technology (CSET) is a policy research organization within Georgetown University's Walsh School of Foreign Service. CSET is concerned with providing data-driven research at the intersection of security and technology, and nonpartisan analysis to the policy community.

At CSET, I worked primarily in the data team on a research project focused on using weak supervision techniques to augment labeling large amounts of training data for machine learning purposes. My research project utilized Snorkel - a data labeling module - leveraging its labeling functions framework to build a weak supervision model. This model was deployed using Apache Beam on Google Cloud Platform on about 100GB of data consisting of paper titles and abstracts. The model final model produced an accuracy of about 90% and was used in various other projects at CSET.

In summary, I

  • Built and deployed a weak supervision model using python Snorkel to automatically label research papers for relevance in a vehicle detection and autonomous vehicles data corpus.
  • Trained the model on 100GB of data using Big Query and Apache Beam, resulting in a 90% model accuracy.
  • Wrote very detailed weekly documentation for the different experiments I tried while building the model.