Richard Mathews II
Projects Blog Book Notes Resume Contact About
YOLBO - You Only Look Back Once
YOLBO - You Only Look Back Once

As part of my CS minor at UT Austin, I conducted computer vision research. My focus was to develop an algorithm that uses space-time information to improve object detection in videos.

PythonComputer Vision
Optimizing Walking Routes for Safety at UT Austin
Optimizing Walking Routes for Safety at UT Austin

When I was an undergrad at UT Austin, I built an algorithm called roUTe designed to help students navigate crime in the campus area. It placed first at Convergent's Spring 2020 Demo Day.

Python
Two Player Chess
Two Player Chess

This is a 2 player chess program playable through command line interface. There are also unit tests to validate the functionality of the program.

Python
Generating Product Improvement Ideas Using LLMs
Generating Product Improvement Ideas Using LLMs

A demonstration of how topic modeling can reduce the costs associated with GPT3.5 requests for a system that analyzes customer reviews to generate product improvement recommendations.

PythonGenAI
Teaching Language Models the Socratic Method
Teaching Language Models the Socratic Method

I conducted research with two collaborators to investigate whether teaching smaller language models the Socratic method could improve their complex reasoning abilities. Our results demonstrate that this approach is highly effective, leading to accuracy gains of 20-40% compared to existing pre-trained language models.

PythonGenAI
WriteAssist - A GenAI Assitant for Writing-Heavy Classrooms
WriteAssist - A GenAI Assitant for Writing-Heavy Classrooms

Reduce teacher burnout and improve student ​writing proficiency by equipping teachers with a ​classroom-aligned AI assistant that automates ​feedback generation and supports students with ​1:1 counseling.

PythonGenAIMLOps
Richard Mathews II