The Internet is filled with videos that show how to do anything imaginable, from using Photoshop to changing a bicycle tire. However, these videos are not as effective as they could be. Research shows that revealing the causal relations between individual tasks and higher-level concepts can enhance how-to videos. We use learnersourcing (collecting useful information from learners to be used by the system) to gather information on the classification of individual steps into concepts. The research has focused on software demonstration videos (e.g., Photoshop tutorials). I intend to build on the existing framework and develop a generalized user interface framework for learnersourcing that will support procedural tasks in general.

I worked at Microsoft on the Office 365 user experience team where I redesigned the product’s setup workflow to make it easier for users to set up the product and reduce the money being spent on support calls. I also interned at Palantir Technologies where I worked on the usability and design of a new web-based version of their core product.