I'm Jon, another programmer on another slice of the internet. Probably solving another problem, or creating another thing. Been programming for 5 years professionally, and will do another, in a loop. Take a look around at some of my stuff, or just say hi.
Some frontend focused projects....
My first large project with a framework. It was a challenge project from the website frontend mentors. Decided to use Next.js with TypeScript and Tailwind for this one.
This is the first edition of my portfolio that landed me the front end engineering role I have now. The framework of choice for this one was Svelte, my personal favorite. Most of the links are outdated by now.
A simple marketing page that was used to practice the fundamentals of HTML and CSS. Another great project from frontend mentor, which I frequently use to stay sharp and up to date.
Some backend focused projects....
This is a fullstack application in Rust, my favorite language. Tried to go as minimalistic as possible, and had a ton of fun. Was hosting this on an amazon ec2 instance, but cost was not worth it.
Wanted to explore PHP and Docker. PHP is an appealing scripting language for backend applications, specifically APIs that service web applications. Docker is a great modern tech that simplifies the deployment of applications.
The stock market is full of data that can be used in various ways. This particular project uses a Rust Api to get data and transform it into views. Got a chance to use Bun.js with TypeScript.
Some open source focused projects....
A simple auth server in Rust. Still a work in progress, but functions really well as is. Built for MySQL and SurrealDB databases for now. Was a good project to understand how authentication works at a low level.
Alpaca is a service that provides endpoints for stock market data. They do not have an official Rust SDK, so I scrapped one together.
I made a simple technical analysis indicator library in Rust. It uses various algorithms to generate technical indicators which are stock market signals to buy and sell.
Some machine learning focused projects....
A tabular binary classification project using the classic titanic dataset. This was my first attempt at tabular binary classification, and I am happy with it. Used the Burn crate from the Rust ecosystem and submitted to Kaggle.
An attempt to train an NLP to classify patent types. The training did not create an accurate model honestly, but was a great experience using the Transformer neural network architecture.
This is the case study of the classic MNIST dataset which contains 70,000 images of handwritten numbers. This was a good lesson for creating a neural network from scratch, as well playing with custom loss functions.
