PORTFOLIO PROJECTS

A Google Chrome extension for enhancing the functionality and user experience of LCR (Leader and Clerk Resources) for The Church of Jesus Christ of Latter-day Saints. Fixing bugs and adding quality-of-life features.

A fast, simple desktop app to help optimize the use of materials by planning cuts based on user-provided specifications. Features clean PDF exports. This app was commissioned and is now free to use.

From the Icelandic words skel (“shell”) and heim (“home”), meaning “Shell Home.” This was built to help me personally in my development projects. I published this to demonstrate proficiency in shell scripting and in hopes that others could benefit from it.

A brief course intended to introduce non-programmers to the Python programming language and data wrangling. Also, a demonstration of network optimization, PDF creation in Python, and a simple Monte Carlo simulation.

Boilerplate code for quick implementation of a documentation site with automatic scaffolding based on folder structure—all built on markdown files.

This site! A project to help me improve my frontend web development skills, advertise any freelance work, and show off some portfolio projects. All of the code is public on GitHub!

A transformer made in PyTorch to handle langauge translation. This project was for a BYU IS Masters-level class I took and was meant to be more educational than practical.

An example serverless web application built with the AWS Cloud Development Kit for a BYU IS Masters-level class. Is built to be production-ready with just a few minor tweaks.

An example project demonstrating serverless deployment in AWS Fargate for a .NET application using multi-stage Terraform deployment. Created for a BYU IS Masters-level class.

Named after the Icelandic word tölum ("speak"), Talum is an iOS app designed to help people learn Icelandic. Created as a school project to showcase app development ability, it is primarily a demonstration and may never be completed.

A pre-trained Text-to-Text Transfer Transformer (T5) fine-tuned for the specific task of converting normal text into KJV-style English. This project was for a BYU IS Masters-level class I took and was meant to be more educational than practical.

A project to learn building transformers for image classification. The model was improved from 10% (random) accuracy to over 80%. This was created for a BYU IS Masters-level class as an educational exercise.

An educational BYU IS project exploring diffusion models and Stable Diffusion. Includes a from‑scratch U‑Net and experiments with Hugging Face's UNet2DModel and VAE-based latent pipelines for text-to-image generation.