Process
Research and Challenges
To research this project, we reviewed various AI chatbot applications such as ChatGPT, Claude, and others to see how they handled parsing large portions of data. Along with this, we interviewed and surveyed employees to find what services would be most valuable to them in their day-to-day workflow.
The biggest challenge with this project was figuring out how to succintly display a wide variety of data types in a simple chatbot. We also were asked to go from zero to MVP on this project very quickly, so there were a lot of discussions to figure out how many components we would launch with. In the end, we settled on a simple text chatbot that could easily be expanded upon over time.
Design and Development
The chatbot's design went through a few iterations before landing on this full-screen, minimal design. We considered a chatbot helper that lived in the bottom right corner of the screen (similar to what you see on many websites today), but felt that having this as a standalone application would better suit our needs.
With this full-screen view, we were able to build something that can be easily expanded with more components, but it also allowed us room to be more creative with these other components. Many of these components are currently under active development.
