Passionate about solving problems, dedicated, hard working, and creative.
Let's build something amazing together!
This page is a collection of projects I've taken on over the years, some of which are still ongoing. Some are closely related to my time at university, and others are purely for my own interest and enjoyment.
2021 - 2022
I'm fascinated with 3D motion systems, so building a laser engraver had been on my todo list for a long time. My CNC Router project taught me a lot, but also exposed some design faults which make it difficult to develop further. I'd been on the lookout for an excuse to start again with a blank canvas. Fortuitously, my mom was looking to expand the branding capability of her boutique leather products business, Patina. This presented the perfect opportunity for me to design and build a machine that would be in regular use, even once my curiosity and excitement had passed. I thoroughly enjoyed exploring the design constraints, learning about various commercial offerings, manufacturing custom components, assembling, and finally tuning the machine. It is still in regular use, with custom artwork and branding now a regular addition to Patina's products.
2019
For my BSc(Eng) Mechatronics final year project, I presented an end-to-end workflow for generating learned concentration estimation models from synthetic aperture radar data, trained on existing passive microwave data. A novel objective function was introduced to account for uncertainty in the passive microwave measurements, which can be extended to account for arbitrary sources of error in the training data, and a recent set of in situ observations was used to evaluate the reliability of the chosen passive microwave concentration estimation model. I was honoured to receive the award for Best Machine Learning Project in the Electrical Engineering Department. The project resources and final report are available on GitHub, at the links below.
2019
The aim of this project was to develop a neural network capable of predicting the position (x, y, z) of fingertips in a pair of stereoscopic images. These images were obtained from a Leap Motion Controller, using their API to create a custom labelled dataset. Each pair of images was annotated with the coordinates of each of the features to be tracked. The full project is available on GitHub, and includes the project report. Have a look at the links below for more information.
2018
In my third year at UCT, I had to build a small robot capable of detecting and following a line marking out a track. The class was ranked according to the time taken to complete the course, and my robot achieved second place. No existing sensors were allowed, so all electronic hardware had to be designed and built from scratch. Here are some photos of the finished product, and some close-ups of the sensors.
2017 - Present
For a number of years I have been working on a CNC router which I designed and built in my spare time. This project has allowed me to indulge my passion for building, as well as to further my understanding of a variety of machining processes, electronics design practices, and CAD & CAM workflows. The project is currently ongoing, but is able to mill aluminium reliably, which is in line with my original goals for the project. Here's a photo taken just after I installed an LED ring to illuminate the cutting area.
2016 - 2017
Towards the end of 2016, I took on my first custom motorcycle build project. A friend had an old Honda XL185 gathering dust on his farm, and wanted to see what it could become. This lead to a full strip down and rebuild of every component - engine, wiring harness, suspension, you name it! I learned so much during this project, and it's definitely something I look forward to doing again. Here are some photos from along the way.
2015
I started becoming very interested in microcontrollers and more advanced electronics towards the end of high school. During my matric year, I designed and built a quadcopter based on an Arduino platform. I particularly enjoyed writing a controller application for my laptop which sent commands to the quadcopter. This allowed me to experiment with a wide range of user input devices, such as PlayStation controllers and LeapMotion hand tracking.