Learner voice before labels
The starting point is what the learner says matters, not a categorical score.
MentorHub explores a different kind of mentoring workspace: one designed to help people listen with intention, translate reflection into action, and return to what matters.
Mentoring conversations can be meaningful in the moment, yet difficult to revisit. When the next session arrives, what was important to the learner? What did each person agree to do? What new evidence should change the plan?
The design challenge is to make that continuity easier, without turning a human relationship into another compliance dashboard.
The prototype centers one coherent journey. Rather than building a large feature catalog, it makes preparation, conversation, commitment, and follow-through visible in the same place.
Understand the learner's goals and the last commitment.
Capture the learner's priorities in their own words.
Agree on a responsible next action, owner, and date.
Review evidence and follow-through, then adapt.
In the live demo, a completed practice check-in creates a record and an accountable follow-up. Closing the action also updates the learner history.
The system is intentionally designed to support a mentor's judgment and a learner's choices, not replace either one.
The starting point is what the learner says matters, not a categorical score.
A check-in never increases progress automatically. Adjustments need explicit evidence.
Every practice session creates a follow-up with ownership, a date, and history.
Fictional records, no student database, and visible limits to what the prototype can claim.
The project originated as an earlier mentoring application. The current portfolio revival keeps the original source history while using a separate branch for an isolated, fictional-data experience.
Original Next.js and Supabase-era ideas preserved as source material. The historical authentication and data workflows were not presented as a finished product.
Recentered the interaction around learner context, self-reported perspective, evidence, and one shared next action.
Next.js, React, TypeScript, custom responsive CSS, GitHub build checks, and Vercel previews. No real learner data is used.
These distinctions are deliberate. The example learners, status labels, and metrics are invented. Practice notes live in temporary browser memory and disappear on refresh. The interface has not been validated in a school or mentoring program.
No real student data, authenticated roles, or connected institutional database.
Mentor discovery, usability evaluation, and evidence of effectiveness have not been completed.
Accessibility reviews, browser testing, security updates, and backend audit are required before any broader release.
Select a fictional learner, load a practice example, record a next step, then complete its follow-up. The best way to understand the design is to experience its continuity.
Launch the MentorHub demo