Student • Researcher • Developer
I am a driven individual at the intersection of computer science, statistics, and finance. With experience in academia, tech, and finance, I bring a diverse perspective. At the moment I am working for Dr. Deen Freelon to analyze the impact of social media data (and the lack therof) on academic research. Furthermore, I am working as a consultant in Oliver Wyman's financial services practice this upcoming summer.
Here are a few things about me
I am a junior at UNC Chapel Hill studying computer science and statistics. At UNC, I run a non-profit consultancy doing pro bono work for healthcare companies. I also am the co-founder of UNC Pokerbots (uncpokerbots.com) where we recently had our first competition with 8 teams (two of which beat MIT pokerbots winning algorithms!!!)
I have 3 years of python programming experience in industry and research. I have skills in Python, Java, C, and R... and a million other random dated languages that classes make us use 😵💫. I genuinely love analytics and statistics. Going all the way back to highschool, I wrote code for Davidson-Davie Community College doing sports analytics for their basketball team. More recently I worked on grant for the NSF to make efficient text extraction models.
With years of research experience in finance and academica, my passion is using programming and statistics to create quantitative societal insights (while not losing sight of the qualitative concerns). I worked at Discovery Capital Managment tracking market trends through sentiment data. This experience led to me working for UPenn Presidential Professor Dr. Deen Freelon on computational sociology research. I have been in the computational sociology space for 3 years now so ask away.
colbyeagan6@gmail.com
+1 980-622-3389
Skills
Data Science
Software Development
NLP & ML
Research
Languages
Python
Java
C
• Working on financial services and quantitative analytics teams to help major financial institutions with credit, liquidity and capital planning, stress testing, regulatory response work, and market risk
• Funded by the National Science Foundation to spend summer of 2024 researching Natural Language Processing and Large Language Models under data science author Dr. Jeff Stanton
• https://uncpokerbots.com
• Directing operations of two separate analytics teams to ensure optimal results for clients
• Acquired partnership with multiple 9 figure non-profits through effective marketing and pitches
• Statistics and data science director managing team of 10 analysts to work for 9 figure nonprofits
• Increased nonprofit donations by 100k after leading analysts to feature engineer 20k rows of data and utilize inferential statistics to pinpoint outlying donor data
• Founded Biokind.org chapter at The University of North Carolina and transformed into functioning non-profit with over 20 members
• Contributing as author and research engineer on computational sociology publication with presidential and distinguished professors Dr. Deen Freelon and David Lazer
• Engineered yaml files to execute multi-tenant cloud pipeline scripts, reducing testing down time by 80%
• Created CI/CD optimizers to run only required scripts on files by analyzing dependency architecture and eliminating all unnecessary hardware costs; reduced hardware costs of DevOps by 30 percent
• Reconfigured solution scripts to execute by regular expression conditionals, creating efficient multi-tenant use-cases
• NLP pipelines implemented for insights in 5m portfolio subsequently reducing manual research by 90%
• Fine tuned sentiment analysis with NLTK and Vader to summarize GET requests in financial context
• Relevant Coursework: Data Structures and Algorithms, Research in Computer Science, Discrete Math, Foundations of Programming, Web Development, Linear Algebra, Methods of Data Analysis, Multivariable Calculus, Computer Systems Fundamentals (C)
• Involvement: Biokind.org Founder and Data Analyst Team Lead, Product Management Club PM, Pearl Hacks Mentor, Poker Bots Founder and Director
Here are a few projects I've worked on. Click on the images to learn more about my work!
AFTSA is a suite of tools for calling APIs in precise time intervals, in statistically consistent time series formalities. AFTSA has over 30,000 downloads demonstrating its widespread adoption within the developer community.
Lexi Graph is a web application to provide people with dyslexia a accesible source to financial information. It was built with HTML, CSS, Javascript, and Flask.
Created a Gaussian Elimination calculator for reducing multivariate systems of equations. Implemented using a GUI for ease of use.
Created executable file Akari software game. built using model view controller design pattern with javafx and maven to assist in building and gui.
Created an applied math web app which generates selected random distributions of data and can fit a line of best fit of any order to the data. Uses simple numpy matrices and plotly for vizualizations, otherwise all the mathematical computations of regression and best fit are done first hand with programming logic.