← Adam Jarick
PRJ / 02

AI Co-Pilot.

Honours thesis. A Chrome extension turning learning analytics into personalised recommendations.

Reads quiz scores, content engagement time, skipped lessons and link clicks from SCORM and Moodle, then routes those signals through an LLM to deliver tailored next-step recommendations directly inside the student's LMS. Demonstrated lift in engagement and module completion.

GitHub
JavaScriptChrome APIsREST APIsSCORMMoodleLLMLaTeX

How it works.

The flow
Inputs

The signals we read.

01

Quiz performance

Per-section scores, miss patterns, and answer-time per question.

02

Engagement time

Time spent on each piece of content compared to the per-lesson target.

03

Lesson skips

Which sections were jumped past versus completed end-to-end.

04

Link clicks

Whether linked readings, case studies, and external references were opened.

05

Quiz retry attempts

Number of attempts before passing a quiz and the score gradient between tries.

06

Scroll depth

How far down each lesson page the student actually scrolled.

07

Video pauses & rewinds

Where students paused, rewound, or sped up embedded video content.

08

Bookmarks & notes

Which items the student saved or annotated for later review.

09

Session pacing

Length and frequency of study sessions across the week.

10

Drop-off points

Where in a lesson the student tends to disengage or close the tab.