Connor Bernard

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2025-127

May 21, 2025

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2025/EECS-2025-127.pdf

Traditional Learning Management Systems (LMSs) often lack the infrastructure necessary to support mastery learning, a pedagogical approach that emphasizes conceptual understanding and flexible assessment timelines over rigid deadlines. This limitation creates significant barriers for educators seeking to implement mastery-based grading policies. To address this gap, we present GradeSuite, a microapp-based LMS designed specifically to facilitate mastery learning implementation while integrating with existing educational infrastructure.

GradeSuite comprises four primary microapps: GradeView, a sophisticated dashboard providing granular concept-level feedback; Concept Map, offering interactive visualizations of learning progression; GradeSync, enabling automated grade synchronization across multiple LMSs; and Instructor Dashboard, delivering comprehensive analytics for course staff. This architecture allows GradeSuite to aggregate and analyze grade data from diverse sources, providing both students and instructors with detailed insights into conceptual mastery while significantly reducing administrative overhead.

We deployed GradeSuite in UC Berkeley's non-majors computer science course --- CS10: The Beauty and Joy of Computing (BJC) --- across two semesters, serving 218 students total. Our results demonstrate substantial improvements in both educational and operational outcomes. Students actively engaged with the platform, averaging 18-19 uses per student per semester, with the majority reporting that it helped them make informed decisions about retaking exams and identifying knowledge gaps. Interviews with course staff revealed that GradeSuite reduced grade management time by 90% (from 10 hours to 1 hour weekly) while eliminating final grade calculation errors. The automated grade synchronization provided by GradeSync significantly improved grade update frequency and accuracy, addressing a key student concern.

This report details GradeSuite's development, architecture, and deployment strategies, including its evolution from cloud to bare-metal and back to cloud infrastructure. The project's success is supported by a robust organizational structure that grew from an individual initiative to a team of over twenty contributors, managed through a carefully designed hierarchical framework that promoted both accountability and open communication. We also discuss future developments, including AutoRemind, an intelligent notification system designed to support flexible deadlines and students continuing work beyond the traditional semester. Our findings suggest that GradeSuite provides a scalable and effective solution to support mastery learning.

Advisors: Dan Garcia

---

BibTeX citation:

@mastersthesis{Bernard:EECS-2025-127,
    Author= {Bernard, Connor},
    Editor= {Garcia, Dan and Fox, Armando},
    Title= {The Development and Management of GradeSuite: A Microservice LMS for Mastery Learning},
    School= {EECS Department, University of California, Berkeley},
    Year= {2025},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2025/EECS-2025-127.html},
    Number= {UCB/EECS-2025-127},
    Abstract= {Traditional Learning Management Systems (LMSs) often lack the infrastructure necessary to support mastery learning, a pedagogical approach that emphasizes conceptual understanding and flexible assessment timelines over rigid deadlines. This limitation creates significant barriers for educators seeking to implement mastery-based grading policies. To address this gap, we present GradeSuite, a microapp-based LMS designed specifically to facilitate mastery learning implementation while integrating with existing educational infrastructure.

GradeSuite comprises four primary microapps: GradeView, a sophisticated dashboard providing granular concept-level feedback; Concept Map, offering interactive visualizations of learning progression; GradeSync, enabling automated grade synchronization across multiple LMSs; and Instructor Dashboard, delivering comprehensive analytics for course staff. This architecture allows GradeSuite to aggregate and analyze grade data from diverse sources, providing both students and instructors with detailed insights into conceptual mastery while significantly reducing administrative overhead.

We deployed GradeSuite in UC Berkeley's non-majors computer science course --- CS10: The Beauty and Joy of Computing (BJC) --- across two semesters, serving 218 students total. Our results demonstrate substantial improvements in both educational and operational outcomes. Students actively engaged with the platform, averaging 18-19 uses per student per semester, with the majority reporting that it helped them make informed decisions about retaking exams and identifying knowledge gaps. Interviews with course staff revealed that GradeSuite reduced grade management time by 90% (from 10 hours to 1 hour weekly) while eliminating final grade calculation errors. The automated grade synchronization provided by GradeSync significantly improved grade update frequency and accuracy, addressing a key student concern.

This report details GradeSuite's development, architecture, and deployment strategies, including its evolution from cloud to bare-metal and back to cloud infrastructure. The project's success is supported by a robust organizational structure that grew from an individual initiative to a team of over twenty contributors, managed through a carefully designed hierarchical framework that promoted both accountability and open communication. We also discuss future developments, including AutoRemind, an intelligent notification system designed to support flexible deadlines and students continuing work beyond the traditional semester.  Our findings suggest that GradeSuite provides a scalable and effective solution to support mastery learning.},
}

EndNote citation:

%0 Thesis
%A Bernard, Connor 
%E Garcia, Dan 
%E Fox, Armando 
%T The Development and Management of GradeSuite: A Microservice LMS for Mastery Learning
%I EECS Department, University of California, Berkeley
%D 2025
%8 May 21
%@ UCB/EECS-2025-127
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2025/EECS-2025-127.html
%F Bernard:EECS-2025-127