Back
Redesigning a Machine Learning Platform for K-12 Students
Making machine learning intuitive for K-12 classrooms.
ROLE
PROBLEM
A new machine learning platform designed for K-12 classrooms was creating significant friction for its users due to core usability issues. Students and teachers were encountering recurring obstacles, hindering the learning process and leading to low task completion rates, which created a significant barrier to the platform's educational goals.
RESULTS
Increased task completion rates by 15%
Redesigned core components to solve recurring usability issues
Identified key usability improvements to enhance the K-12 user flow
Project Results
CODAP Plugin for students to use Machine Learning components in their datasets
Heuristic Evaluation
Looked at current platform to find user pain points and how to address them in CODAP Plugin.
Why CODAP?
Low Fidelity Prototyping
High Fidelity Prototyping
Component Library
Design Changes
Demo of CODAP Plugin
Final Outcome
The platform's recurring usability flaws were creating a frustrating experience, so I redesigned core Figma components to directly address these problems. This component-level fix enhanced the overall user flow and led to a measurable 1R% increase in task completion rates. By identifying and solving these usability gaps, we made the complex machine learning platform more intuitive and effective for its new K-12 audience.
Takeaways and Next Steps
Adding remaining Scientific Inquiry and ML components
Conduct Usability Testing
Using CODAP API to launch usable public plugin