Full-stack platform where neurodiverse college students reflect on their teamwork with an AI coach, and instructors see engagement analytics and AI-generated insights.
Python LlamaIndex AI Agents RAG
Highlights
Students reflect on their teamwork with an AI coach that replies with interactive widgets, like rating the strengths it noticed in a reflection.
The coach identifies each student’s strengths and uses retrieval-augmented generation (RAG) to recommend tools for the challenges they describe.
A strength-advising agent turns the student’s VIA survey and conversation into a strength profile and an action plan they can edit.
Instructors get engagement analytics and AI-generated insights that tag each reflection with the strengths and challenges it shows.
Developed with SFSU psychology and education faculty to put their research into practice.
Gallery
Select any image or video to view it full size.
Fig. 1 Weekly reflection: the coach names the strengths it noticed and asks the student to rate each one, then suggests tools for their challenge.
Fig. 2 Strength advising: the agent affirms a strength from the student’s story and builds their strength profile and action plan alongside the chat.
Fig. 3 Instructor analytics for the week: enrollment, positive and negative reactions, AI interactions, and time spent per student.
Fig. 4 Instructor insights: each reflection summarized and tagged with its strengths and challenges.
Fig. 5 Instructor view of a student’s strength profile and action-plan progress.