Cole Harper

Backstory:

One of my really good friends told me about their LED Matrix that they were playing around with.

Idea:

I love football, especially college football. What if I could create my own scoreboard that would automatically cycle through all football games that are going to happen during the week. Then on game day I can see live updates.

Implementation:

So using the rpi-rgb-led-matix library on a pi to drive the actual LED, I built out a dockerized go solutions that utilizes Redis for caching game data, and then a native python process to drive the LED rendering.

Dockerized Services:

• Data collector: ESPN API integration & data normalization - Golang
• Queue Manager: Display prioritization & event detection - Golang
• Redis: Central data store & message bus

I chose to use Golang because of its concurrency ease.

Non-Dockerized Services:

• Display Manager: LED rendering & visual presentation - Python
  • This had to be separate as it inherited the Python bindings from rpi-rgb-led-matix library

Future Enhancements:

What currently drives this is a Raspberry Pi. I want to change this to be a ESP32. The advantage of this is the ESP32 can use dedicated hardware peripherals (like I2S + DMA) to clock out pixel data at precise, deterministic timing without CPU involvement, whereas the Pi relies on software-driven GPIO that gets interrupted by Linux's non-real-time kernel.

• Refactor to use a ESP32.
• Implement other modes such as a clock, weather, etc.

2024

Advent of Code is a series of programming challenges that serve as a virtual advent calendar leading up to Christmas. This year, I participated in the challenge and successfully completed all the problems. It presented a stimulating and diverse range of computer science questions, making it an enjoyable experience. Day 5 stood out for me as it provided an opportunity to implement algorithms that I had long desired to explore. I completed all of the challenges in Go, however next year I am planning on tackling them in either Rust or Swift. I highly recommend participating in the challenge next year.

2025

I completed this year's answers in Python. All of the problems this year felt approachable... Well except for day 10...

I created this website as a personal project to display my portfolio and learn how to build a complete application. As you can probably guess, I’m not a front-end developer. I’ve gained a newfound respect for them! This website has taken longer than I’d like to admit to get working! But it’s been a great experience! The front-end is built with Svelte and Tailwind and I have PostgresSQL servicing all data on the frontend. These services are Dockerized and running on a Digital Ocean Droplet with a SWAG (Secure Web Application Gateway) container managing all the networking. I wanted to be able to experience the full flow of architecting, coding, and testing a small project, and I’ve learned a ton along the way! This website is still a work in progress, but I’m happy to share it with you! I plan to keep iterating on it and eventually make the GitHub repo public.

This was a project that I worked on for an iOS class at the University of Texas. We were required to construct the application from the ground up. The objective was to develop a habit tracking application that utilized virtual animals to provide incentives for building positive habits through interactions with the virtual animals. We developed this application over the course of a semester. We utilized UIKit and SwiftUI for the user interface and CoreData for persistent data storage. Additionally, we employed SpriteKit for rendering the animals. Overall, this project provided an enriching learning experience.

I created a small private Discord bot that would send memes and other inside jokes when someone in the server sent a message that contained a keyword. Modes could be turned off and on, new keywords added, all live from Discord. This project was exceptionally fun to do, but it was done using the Discord.py library, which I felt was difficult to use. I have since decommissioned the project, but I have plans to create a new Discord bot written in Go instead.

This was a TAMU 2020 Hackathon project where I was on a team of 4 that came up with a proof of concept of a website that provides drivers with an easy-to-use, interactive map, depicting all possible roads considered hazardous in a given area. Based on previous crash data, our website provides predictions of what percent of the time a certain road will be considered "dangerous". Since all the information on our website is neatly presented and easy to understand, users are able to make more informed decisions about what route is safest to travel given certain conditions. The predictions were provided by a Microsoft Azure model.

This was a 2020 HackTexas Hackathon project where we developed True Colors, which reveals the parent company and its associated ESG (Environmental, Social, and Governance) ratings, enabling users to uncover potentially hostile companies behind everyday goods. By removing an enormous barrier to voting with your wallet, this smooths the path to backing companies providing for more social good. ESG ratings calculated by several different firms track everything from pollution and waste to product liability and human capital, as well as corporate behavior. The app was written in React, Node.js, and SQLite.