Node Learn
How it worksFor studentsFor teachersPhilosophyFAQAboutGet started

How it works

One goal, one relationship, a visible chain of work.

There is no catalogue step and no matching quiz. The loop below is the entire product — everything else exists to support it.

LEARNING RELATIONGoalTasksEvidenceFeedbackChangerevise

For the learner

The student loop

  1. Describe a learning goal

    Say what you actually want to be able to do — in your own words. No placement quiz, no catalogue to browse first.

  2. Find or invite a teacher

    Search by topic and read self-described profiles. Send a short request explaining your context, or accept an invitation.

  3. Establish a learning relation

    When a teacher accepts, you get a shared workspace. That relation is the container for everything that follows.

  4. Agree on goals and next tasks

    Break the goal into a small number of concrete tasks you both can see. One next step is always clear.

  5. Submit evidence, get feedback

    Share the actual work — a file, a write-up, a link. Feedback attaches to the evidence, so revisions have a visible history.

  6. See a change you can explain

    Progress is recorded as a short, explainable proof: what you can now do, and the evidence chain behind it.

For the teacher

The teacher loop

  1. Write a concise teaching profile

    Describe your experience, method and the topics you take on. Self-described — the platform never pretends to have vetted you.

  2. Receive a relevant request

    Requests arrive with the learner’s stated goal and context, so you can judge fit before replying.

  3. Review and accept or decline

    Declining is a normal, low-friction action. There is no response-rate score punishing you for saying no.

  4. Set goals and tasks

    Turn the goal into a short task list in the shared workspace. Both sides see the same structure.

  5. Review evidence and give feedback

    Comment directly on submitted work. Your feedback becomes part of the learner’s traceable record.

  6. Track who needs attention

    A quiet workbench shows active relations, pending reviews and which learners are waiting — no vanity dashboard.

Vocabulary

Six words worth knowing.

The product uses these terms precisely. They are the data model as much as the interface.

Knowledge node
A single idea or skill. The first-class object in the system. A "course" is only a temporary path between nodes.
Learning relation
The durable link between one learner and one teacher. It holds the goals, tasks, messages, evidence, feedback and proof for that pairing.
Goal
What the learner wants to be able to do, written in their own words and agreed by both sides.
Task
A concrete next step inside a goal. Kept few and visible so there is always one obvious thing to do.
Evidence
Actual work — a file, a write-up, a link. Feedback attaches to it, so revisions leave a visible history.
Proof
A short, explainable record of a change in capability. Every proof traces back to the evidence behind it. No mystery scores.

Messaging

Messages carry structure, not just text.

Inside a relation, a message can be a goal proposal, a task, a submission, a piece of feedback or a milestone. That means the important moments stay findable instead of scrolling away.

  • GoalBe able to explain recursion well enough to teach it.
  • TaskWrite a recursive tree walk and describe each call in plain English.
  • Evidencetree-walk.py + a 200-word explanation
  • FeedbackThe base case reasoning is solid. Your explanation skips why the stack unwinds — try again on that paragraph.
  • ChangeCan explain recursion including stack unwinding — 2 evidence items, 1 revision.

Illustrative example, not real user data.