StreamKit is a web-based app for quickly onboarding datastreams and analytic capabilities for a national lab.
ROLES: UX Designer & Visual Designer
PROBLEM: The user needs to set queries and parameters for a specific machine learning based anomaly detection algorithm without in-depth knowledge of how the algorithm works.
PROCESS: I first met with the analytic developers to understand the inputs to and outputs from the algorithm and what form they take. I then created initial mockups to help our intended analyst understand how to interact with the analytic.
The mockups below show primarily the setup process for the alert anomaly analytic. As the main application is for streaming data, the alert may not fire for some time. Because of this, we want to use the setup process to look at a snapshot of previous data and show which alerts would have occurred to give the user an idea of what alerts may be generated in the future.
I also generated figures for use in stakeholder presentations to assist in explanation of the Stream Kit system.