Career Path Recommender: Student Assessment Platform
Student assessment platform that explains career recommendations and gives administrators a separate view of institutional results.
- Education
- Full-stack developer, assessment model through dashboards
- AI Systems
- Deployment offline
What problem does Career Path solve?
Career guidance in schools does not scale cleanly. One counsellor may cover hundreds of students, paper assessments move slowly, and results can arrive after the decision they were meant to inform.
How does Career Path work?
Career Path turns a structured assessment into recommendations with visible reasoning. Students receive a result they can question, while administrators see a separate view of cohort choices and institutional records.
Which architecture decisions shaped the build?
- Recommendations shown with their reasoning
- A career suggestion a student cannot interrogate is not guidance. Showing what drove the result makes it something to discuss with a counsellor.
- Document store for assessment responses
- Assessment shape changes between institutions and cohorts. A document model absorbed those revisions without a migration for each one.
- Institutional view kept separate from the student flow
- Administrators need aggregates, students need one result. Separate surfaces kept both readable.
Field notes from the build
Deployment status. The previous hosted demo returned 404 when the links on this site were last checked on 15 August 2026, so no live link is offered here. The source repository is public and the screenshot is from the working build.
Career guidance becomes difficult when one counsellor is responsible for hundreds of students. The platform handles the structured assessment so the human conversation can begin with evidence already on the table.
A recommendation must explain itself
The result shows which answers influenced the recommendation. A student can examine the reasoning, challenge it and take a more informed conversation to a counsellor.
The data model accepts changing assessments
Institutions revise their questions between cohorts. The document model accepts those changes without demanding a schema migration for every new assessment shape.
Evidence in the build
An explained recommendation flow, role-based access and a clean boundary between the student result and the institutional view.