John Haas
Prospective Data Scientist
I'm a senior 4.0 Computer Science student with a Math minor at the University of Portland. I am actively looking for a full-time job either as a data scientist or a machine learning engineer. Below is a selection of work and school projects that demonstrate my Computer Science talents.
Research
Contaminants in recycling is a major issue. If recycling has too much contaminants, it must be taken to a landfill. Additionally, machines used to process recyclable materials are often not prepared to handle contaminants, and certain contaminants, like plastic shells, can cause these machines extreme damage.
For the past two years, my research group has been pursuing an internet of things and machine learning solution to eliminating contaminants in Portland recycling. I created a Raspberry Pi hardware setup capable of collecting visual, audio, and location data of material collected on recycling truck routes.
Alongside creating hardware, I also created the software solution. Using sample data, I removed noise and preprocessed the image data. Then, I created a convolutional neural network that would classify objects as one of 6 material classes. The current model classifies images with over 80% accuracy.
Capstone Project
For the past year, I have been working on my capstone project. Prosthetic arms are expensive and can be difficult to use, which is especially bad for children, who are constantly growing. To remedy this, our team was tasked with making a cost-efficient prosthetic arm that is easy to use.
The first portion of this project i was responsible for was creating a computer vision solution for choosing hand grips. Normally, users have to use a special band to map specific nerve pulses to grips that they want to use. This is incredibly difficult for a child, so instead I trained a neural net that would map the images to grips so children do not have to use this complicated piece of hardware.