A learning platform redesign for smarter study planning and in-context AI support
My Role 🧠🎨
As the team’s UX Designer, Project Manager, and Prompt Engineer, I translated research insights into actionable design solutions by embedding learning science principles into product decisions. I also coordinated cross-functional collaboration and developed AI prompt strategies grounded in the Socratic Tutor model.
Ashley Xu, Tingyue Cui, Tracy Ciou, Lily Lee, Sherry Li
Overview
This 7-month project was a collaboration with the Open Learning Initiative (OLI), an open-source platform that applies learning science and technology to improve outcomes and expand access to high-quality digital courses.
Our goal was to enhance engagement and learning by improving how students interact with the platform’s AI assistant—DOT (Digital Online Tutor)—and by reimagining the student homepage to surface actionable insights and integrate AI more seamlessly into review and practice workflows.
DOT is trained on the REAL CHEM courseware and functions as a contextual, Socratic tutor. It engages students after difficult topics, offers feedback on selected-response questions, and supports metacognitive reflection.
We approached this goal from three directions:
Redesigning the student homepage to support data-driven study planning with real-time proficiency insights and AI-generated personalized practice
Reworking paragraph-level DOT activation through two rounds of UI redesign and usability testing
DOT appeared as a clickable icon across OLI course pages, offering AI-generated summaries, hints, and answers to student questions. However, early research showed that students often overlooked DOT or found its responses too long or poorly timed, limiting its usefulness during learning.
At the same time, students struggled to plan effectively without clear, actionable insights. While OLI collects rich behavioral and assessment data, the homepage surfaced little of it in a way students could understand or use. Learners lacked visibility into their progress, weak areas, and what to study next—making the experience feel static, fragmented, and disconnected from their actual learning needs.
Outcome
The project involves
AI Prompts Launched
79
+
Contextual prompts implemented in the Summer 2025 RealChem1 course
DOT AI Awareness
125
%
Increased through onboarding and contextual prompt strategies (based on user interviews)
User Testing Sessions
25
+
One-on-one, 1-hour usability sessions to gather actionable design feedback
Homepage Usefulness
84
%
Based on 3 rounds of testing with Summer RealChem students