My version of the solder fume extractor, attempting to bring the best in portability for soldering:

• fume mitigation with the widely-available Weller WSA350F Carbon Filter
• USB PD to power soldering irons such as the TS80P/TS101/Pinecil
• cordless and rechargeable power supply using the widely-available Ryobi ONE+ 18v battery packs
• voltage conversion to power a typical 140mm PC fan
• the Ryobi battery serves double-duty, providing a stabilizing mass as the base of the fan
• a large hinge allows for tilting the fan for optimal positioning the filter enclosure is able to be easily opened, allowing for easy replacement of the carbon filter when needed
• a hexagon grid pattern is used for style and functionality as the mesh pattern, the small hexagons are intended to keep tiny fingers safe

Download the files to make your own fume extractor on Printables.

Over the course of a couple months, I developed an application to demonstrate what warehouse operations would look like when using a Zebra handheld barcode scanner (TC52 and WS50) with the new Vuzix Z100 Smart Glasses.

Initially conceived as a tech demo to validate the concept, the project quickly expanded into a fully-featured warehouse picking application to highlight how the Z100 Smart Glasses could be used to improve safety and speed in the warehouse.

Implemented in Kotlin using the latest best practices, Zebra’s DataWedge API, the Vuzix Ultralite SDK and backed by a state machine to model the picking flow, this app rapidly went through several rounds of iterations between myself, Vuzix’s Software Director, the CEO, and industry experts. The outcome of this collaboration and fast prototyping resulted in a polished demonstration of the Z100 Smart Glasses and was demoed at CES 2024.

Redesigned and simplified the Android Open Source Project’s DocumentsUI app to serve as the new File Browser for Vuzix Smart Glasses, enhancing user experience and maintainability. The project directive was to re-imagine the user interface and make the code easier to maintain in the future. This included removing unnecessary components, like the side navigation drawer, and optimizing the interface to work with touchbar navigation. The project involved over 15,000 lines of code changes, incorporating both Java and Kotlin. As the sole contributor, I oversaw the entire development process, from initial design to final implementation, significantly improving the device’s usability and accessibility. I used a combination of Java and Kotlin for this project, and a large majority of the app was converted to Kotlin over the course of the project.

The New York Times has a puzzle in their NYT Games called Letter Boxed. This program is a solution finder for that game.

This project came about as a reult of nerd sniping myself in trying to find a generic solution for any given Letter Boxed puzzle. I chose to implement the program in Kotlin, using Coroutines for multi-threading.

I ended up using this project source code as case study in the Kotlin training course I hosted while at Vuzix.

The project source is available on GitHub.

After looking for magnetic letters that would hold up to my toddler and being disappointed by what I found, I designed my own.

Using Fusion 360 and the Roboto Mono font, I laid out the position for each magnet and created a void in the models so that strong neodymium magnets could be inserted mid-print.

The magnets are fully enclosed in the plastic, so it’s safe for a child to play with. They can be easily cleaned and won’t scratch the refrigerator or dishwasher surface when slid around.

Check out the project and print your own on Printables.

Developed in Kotlin, I introduced a feature to the Hopper app that allows users to attach frequent flyer details to each passenger on an itinerary and automatically pass this information along when booking flights.

During the development process, I updated the underlying profile models used across various verticals within the Hopper app (Air, Car, and Hotel), transitioning them from Java to Kotlin. This effort was closely coordinated with other teams to ensure minimal impact on the app’s functionality and to mitigate the risk of regression.

The feature was successfully rolled out with a smooth transition, enabling customers to seamlessly complete the checkout process. Now, users no longer need to re-enter frequent flyer details directly with airlines, enhancing the overall user experience.

Added captions support for Alexa’s voice, making the service more accessible to deaf and hard of hearing users. Adding this feature to the AVS SDK enabled third-party device developers to add captions support to their products. This work, written in C++11, is part of the open-source Alexa Voice Service SDK available on GitHub. I was responsible for the design and implementation of the feature, from start to finish. Developing this feature required me to learn and become proficient in C++.

The project code can be found on the AVS SDK GitHub

Community-maintained drivers for the DYMO LabelWriter printers. DYMO no longer supports Linux, so this is a fork of the official DYMO SDK for Linux, updated so that it compiles with CUPS version 2 or higher.

Check out the DYMO SDK for Linux on GitHub.

Badge printing application for Raspberry Pi that logs user information such as name and email and prints a name badge and QR code associated with that information.

Features include:

• DYMO LabelWriter integration
• Google Sheets upload
• NFC reader integration
• full customizability with the administration page

Check out the BadgeHub project on GitHub.

Currently a work in progress, the goal is to be a mead-making helper app for Android.

If you’re intrested in learning how to brew mead, check out the mead subreddit.

This app is being designed to make mead-making easier, from recipie creation, to calculations and reminders about when to perform certain tasks. Let me know if you’re interested in beta testing the app!

Check out the Must Watch app on GitHub.