Connor Pennington
An undergraduate computer science with a passion for Artificial Intelligence and Machine Learning. I thrive on tackling challenging problems and continuously learning and enhancing my skillset.
Experience
Software Engineer Intern | 402 Software Engineering Group - Summer 2024
Leveraged machine learning by implementing a model trained on 1 million PDFs to automate table extraction from over 600 unique formats. Created an algorithm to programmatically reformat the tabular data into XML, XAML, and .cs files. This resulted in a 20% increase in software development productivity by eliminating the need for manual table extraction and reformatting. (Python) (C#) (WPF) (XML) (XAML) (Pandas) (Machine Learning)
Projects
Personal Projects
Climate Crisis Film Repository
Star
This website was created to provide an accessible and informative platform for individuals interested in exploring the climate crisis through film. The website includes a curated film repository, expert analyses, a film recommendation system, and discussion boards. (Python) (LLMs) (API) (Streamlit)
This website was created to provide an accessible and informative platform for individuals interested in exploring the climate crisis through film. The website includes a curated film repository, expert analyses, a film recommendation system, and discussion boards. (Python) (LLMs) (API) (Streamlit)
Lecture Genie
Star
LectureGenie is a software program specifically designed to enhance the learning experience with recorded lectures. It takes a lecture video as input and generates quiz questions, notes, and flashcards to aid understanding. It processes videos with and without audio. (Python) (AWS) (LLMs) (API) (Computer Vision) (NumPy) (Streamlit)
LectureGenie is a software program specifically designed to enhance the learning experience with recorded lectures. It takes a lecture video as input and generates quiz questions, notes, and flashcards to aid understanding. It processes videos with and without audio. (Python) (AWS) (LLMs) (API) (Computer Vision) (NumPy) (Streamlit)
Weak-to-Strong Generalization in AI Moral Reasoning
Star
This project explores the potential for creating superhuman AI systems capable of moral reasoning that can be evaluated by humans, or less advanced AI systems which in turn can be evaluated by humans. The goal was to develop a prompt suite for weak-strong generalization of moral reasoning. (Python) (OpenAI) (Mistral) (Claude) (NumPy)
This project explores the potential for creating superhuman AI systems capable of moral reasoning that can be evaluated by humans, or less advanced AI systems which in turn can be evaluated by humans. The goal was to develop a prompt suite for weak-strong generalization of moral reasoning. (Python) (OpenAI) (Mistral) (Claude) (NumPy)
- Connor Pennington
- [email protected]