Joseph Jay Williams

Joseph Jay Williams

VPAL Research Scientist
Joseph Jay Williams

Joseph Jay Williams's research designs adaptive systems for online content, like intelligent lessons that emulate our favorite teachers in personalizing explanations. This computational cognitive science approach requires integrating his research in psychology & education, human-computer interaction, and statistical machine learning.


To make any static website become intelligently adaptive, Joseph uses powerful systems for randomized A/B experiments. Examples range from adaptive explanations for how to solve math problems, to self-personalizing emails to change people's behavior. His systems continually crowdsource new 'A' and 'B' designs from psychological scientists, using randomized comparisons to evaluate how helpful these alternative designs are, for people with different profiles. Algorithms from statistical machine learning use this data for real-time enhancement and personalization, by changing which designs are presented to future users. By enabling scientific discoveries from each user to directly impact engineering improvements for the next, Joseph's work provides intelligent content that never stops improving.

Contact Information

125 Mt. Auburn Street, 4th Floor
Cambridge, MA 02138

VPAL Research