Experience:
Wisconsin Department of Natural Resources
2019 - Present
A mixture of Python and R scripts were used and automated to create a data processing pipeline which runs and utilizes the output of a pre-trained object detection model (MegaDetector) to automatically classify all images in and added to the Snapshot Wisconsin project. This implementation decreased the human image classification backlog by nearly 10 million images (> 50%) and gives an initial classification to all new images within minutes of upload.
Exceptionally designed data visualizations can positively impact a project in ways which are difficult to explain. Detailed temporal data associated with cameras, laid out in an intuitive way, help lead to the identification and effective communication of issues across the entire project which were otherwise unexplainable. With consistency in understanding of the issues, resolution were quickly able to be decided upon and implemented.
While working in the Office of Applied Science of the Department of Natural Resources, I have taken on the responsibilities and roles of Unmanned Aerial Systems Coordinator and Pilot. The responsibilities include managing a drone flight program, earning and retaining a FAA Part 107 pilot license, and planning, coordinating, and executing flight missions related to scientific research being conducted by fisheries and wildlife biologists and research scientists.
Researchers using Snapshot Wisconsin data often are R savvy users. To better serve these users with consistent, robust, and easy to access data, I designed and created a prototype R package to help pull a single use case worth of data. The prototype had such a positive impact on the Snapshot Wisconsin researchers that I was immediately logging feature requests. Since then, the R package has become a mainstay for researchers.
The Snapshot Wisconsin Data Dashboard is a data visualization application using R and the Shiny package. The purpose is to visualized Snapshot Wisconsin species classification data for public use. Future iterations will be used to help species specialists and species committees make decisions. This was the first successful effort to share Snapshot Wisconsin data with the public on an ongoing basis.