Erik Sundblad

Software Projects and Works

UniTrail

AR driven Campus Navigation Tool

This Mobile app was the subject of my senior project at MSU Denver. The application utilized unity to generate an augmented reality view on a users phone that could navigate to and from specific rooms on campus. The app was developed in a flutter framework utilizing Dart, we hosted a user and room # database through Firebase. The app integrated google maps API for overworld view and received full marks from the department and our professor. Demo Video

Main Application Splash
Unity model for nav
Augmented Reality early phase appearance

Modeling Basil Growth

An independent study modeling the growth of basil through Machine Learning

For this project we collected environmental, physical and chronological data surrounding the growth of a sample of basil plants. We used Arduino boards to collect humidity and temperature each hour. Physical growth was approximated using an overhead camera capturing hourly images. These images were then processed to generate a “green pixel count” that would represent growth of the basil. After aggregating all sample data. We evaluated the growth through linear and polynomial regression models to identify any strong correlations between growth and temperature, humidity, or soil humidity. This project was independent study conducted outside of the class room under the mentorship of Professor Jiang Feng. Code snippets and results can be found in the shared google colab for the project.

Raw Growth Data Distribution
Sample01 relevant histograms
Early polynomial Machine learning Model

GO4 ETL tool

Software designed for Hydrologik LLC as data ETL tool

GO4 is a replacement tool for a legacy software know as “gopher”. The program functions as an ETL tool (Extract, Transform, Load) for Hydrologik LLC. The program collects form AIP’s, and shared data locations on the server. After transforming, the data is stored in an SQL database and some is brokered to the state of Colorado Hydrobase API. The tool is written in Python developed with multiple monitoring breakpoints utilizing SMT email notifications. The tool is hosted and developed using AWS on an EC2 instance.

SME Tracker

Web Application for emloyee progression monitoring

While interning at Dynatrace during the summer of 2022 I and a colleague had the opportunity to develop a tracking tool for employee progression through sets of advanced company training. The tool itself was developed in JavaScript utilizing node express JS for server infrastructure. We utilized a non-relation database for the backend (CosmoDB). The tool was hosted through an Azure app instance and fully integrated Azure Map and SSO authentication. Public source code for the project can be found at github