Caleb McIrvin

I'm currently working through a number of reinforcement learning papers, developing functional solutions to enhance my knowledge of the field.

I developed a quantum circuit visualization tool to parse OpenQASM 3.0 circuits as a contribution to the open-source project QBraid. To ensure code quality, I iterated over multiple self-imposed code improvement versions, emphasizing clarity and simplicity to promote future expandability.

As a competitor for Homeschool Science Colorado, one of the premier Science Olympiad teams in the state of Colorado, I designed, created, and currently maintain their website.

Using Python's cv2 library to perform image transformations / overlay court data on an overhead view of a volleyball match. Adding actions such as serve location, pass type, set location, spike distance, and block type to be used in collecting match data for statistical analysis.

Pulled data from a .csv file of data collected by https://covidtracking.com/race, sorted the data by ethnicity/state, calculated relevant statistics, and displayed the data using an interactive GUI, while using GitHub for version control/collaboration.

Used the PRAW API to download Wyze subreddit posts/comments, ran sentiment analysis on the data using NLTK's VADER library, and uploaded the data for graphical analysis.

Used the Pomegranate library to create a hidden Markov model for POS tagging as part of the Udacity Artificial Intelligence Nanodegree program.

Created an adversarial agent to play Knights Isolation as part of the Udacity Artificial Intelligence Nanodegree program.

Implemented and analysed a forward-planning agent using symbolic logic and classical search to create an agent which uses progression search to solve planning problems as part of the Udacity Artificial Intelligence Nanodegree program.

Created my own image classifier using PyTorch as part of the Udacity AI Programming with Python Nanodegree program.

Developed a sudoku solver to solve regular and diagonalizable sudoku puzzles as part of the Udacity Artificial Intelligence Nanodegree program.

Used a pre-trained classifier to successfully identify different dog breeds from images as part of the Udacity AI Programming with Python Nanodegree program.

Implemented a simple route planner using the A Star search algorithm as part of the Udacity Intro to Self-Driving Cars Nanodegree program.

Created an image classifier that would load and visualize hundreds of images of traffic lights, pre-process the images, extract features from the images, calculate the error, and evaluate my model as part of the Udacity Intro to Self-Driving Cars Nanodegree program.

Translated Python code into C++ code as part of the Udacity Intro to Self-Driving Cars Nanodegree program.

Implemented a matrix class in Python as part of the Udacity Intro to Self-Driving Cars Nanodegree program.

Developed an Android application, Bee Notified, using Android Studio, Java, and XML with the goal of notifying beekeepers when farmers in their area would spray pesticides. To allow the beekeepers to secure their hives against pesticides, we created Bee Secure, an Arduino mechanism that would allow the beekeeper to remotely secure the entrance of their hive against pesticides.