285 – Programmer, Building a Backbone for Multi-Agent Intelligent Tutoring Systems

Project Name
Building a Backbone for Multi-Agent Intelligent Tutoring Systems

Project Description
Over the last few years, a likely solution has emerged: service-oriented design and combining components from multiple Intelligent Tutoring Systems (ITS), which leverage artificial intelligence to speed up learning. The research problems that this work attempts to address are: 1) Multi-Agent ITS Services: Rapid and seamless integration of multiple ITS services that interact like agents in real-time to provide a coherent and effective learning experience. 2) Plug-and-Play Interoperability: Reducing barriers to adding services to ITS down to a level that a class of students could, within a day, add a new agent as a service. 3) Blending Expert Knowledge with Machine Learning: Models that allow experts to explicitly declare knowledge and then learn from data to improve agent performance, without needing to throw out the existing expert knowledge.

Job Description
The goal of this internship will be to program new web services that leverage artificial intelligence, machine learning, and semantic messaging to make it faster to build AI-based services that support learning. The specific tasks will be determined based on the status of the project at the time of the internship as well as your interests. Possible topics include work with: (1) building new tutoring system components and models, (2) machine learning models that identify efficient ways to leverage user data to improve the performance of tutoring system components in real time, (3) running brief usability tests with users of these components with new users of minimal-working-example tutoring system.

Preferred Skills

  • Python, JavaScript, Java
  • AI Programming or Statistics
  • Strong interest in artificial intelligence, machine learning, and human and virtual behavior

Apply now.

Go back to the summer research program list.