Skip to content

Capstones and applied projects Learning and making, one path.

A capstone, a thesis, an applied project inside the participant's own organization, an assignment between modules. In Ulern the learning and the project are one plan under your method, with the AI tutor on every draft and a supervisor at every gate.

The projects

Every programme ends in real work.

Ulern runs the shapes a business school already uses. Each one is a method: steps, what each step must contain, and the gates where a supervisor signs.

Capstones and theses

The final deliverable of a degree or an executive programme: a memo, a plan, a thesis, with a supervisor who signs each gate.

Projects in the participant's own organization

The applied project that makes a programme worth its fee: real data, a real decision, a deliverable the sponsor can read.

Assignments between modules

The work between two modules of a modular programme: a date, a gate and a person at the end, instead of a reading list and hope.

Custom programmes for a corporate client

A client's own track under its own method, in a private space, with a record you can show the client at the end.

The method

Three kinds of step, and no more.

The supervisor designs the method once. Every participant walks it, and every supervisor sees the same shape.

Work

Write a part of the deliverable: the research question, the design, the analysis, the recommendation. What the step must contain sits above the draft, not in a rubric nobody opens.

Decision

Choose a direction and say why. Each option names what it sets up. The AI tutor does not grade a decision, and the page says so.

Gate

Wait for something first: a draft before any AI help, or a supervisor's approval. A gate is never a dead end; it says who holds it, since when, and what can go on meanwhile.

How a project runs

The supervisor sets the method. The participant does the work.

Each step sits in the participant's plan beside the topics it needs. The AI tutor teaches those topics before the participant writes, checks the draft against what the step must contain, and asks for a revision. Nobody learns one thing and makes another, and nobody skips to the end.

1

Method

The supervisor sets the steps.

2

Draft

The participant writes a step.

3

Review

The AI tutor reads it and asks for more.

4

Revise

The participant improves it.

5

Sign-off

The supervisor approves it.

6

Assemble

The finished work comes together, with its record.

Sign-off

Two keys close a step.

The AI tutor's review happens before the work reaches the supervisor. The supervisor's signature is the one that counts.

The AI tutor's review

  • Reads every draft against the step's criteria
  • Tells the participant what is missing
  • Never writes the work
  • Can be accepted with the gaps recorded

The supervisor's signature

  • Counts over the AI tutor's review
  • Returns work with a reason
  • A limited number of rounds
  • Closes the step

No step is skipped without a record, and a supervisor decides at every gate.

What the supervisor sees

The steps waiting on you, first.

Supervision in Ulern is a queue, not a mailbox. The reading and the asking-for-more happen before a draft reaches you, so what you open is ready for a decision.

Waiting on you

Every step that needs your signature, with the participant's draft and the AI tutor's review beside it. One email each morning lists them.

Return with a reason

A returned step reopens with your reason attached, where the participant works. The rounds are limited, and approval is still possible when they run out.

Where everyone stands

Each participant and each cohort by step, with the ones waiting on a supervisor first, and since when. Reports per participant, per cohort and per programme.

The record

Every draft, review and decision, by name and time. If a question about the work comes up later, the answer is on the record.

On real screens

How a capstone runs, in two minutes.

2 min 16 s, captioned, on a demo academy we keep for this; the people in it are invented. Read the walkthrough as a post.
The deliverable

Real work, with the record of how it was made.

Don't ban AI. Don't grade what AI wrote. See how the work was made. The AI tutor never writes a step of the work. At the end it can smooth the joins between sections, and every such change is shown to the participant and needs their approval before anything is exported. Every draft, review and decision is on the record. Ulern does not claim to detect text written elsewhere; it gives the supervisor the drafts, the rounds and the feedback, which is what a judgment about the work stands on.

  • The finished work, assembled from its steps
  • Exported to Word, PDF or Markdown
  • A record of every draft, review and decision
  • The supervisor's rules kept: word limits, format, a transparency note
  • Evidence on the topics the work stood on
Custom programmes

Show your client what their people did.

A client that pays for a custom programme will ask what changed, and a satisfaction survey does not answer that. In Ulern every participant applies your programme to a real challenge in their own organization, and the result comes with the record of how it was achieved.

How a client's space is kept apart · How a client's people are counted

A private space for each client

The client's people, cohorts and projects sit in a space of their own, running your programme. What your school sees of it is agreed up front: nothing, the numbers, or the full record.

A project in the client's own work

The steps, and what each step must contain, are your method. Each participant brings a real challenge from their own organization.

Your supervisors decide

The AI tutor checks every draft against your criteria and never writes the work. Your supervisors return or sign at each gate.

The result, and how it was achieved

Each finished deliverable exports with its record: every draft, review and decision. Reports show each participant and each cohort by step.

Questions

Questions supervisors ask

Who decides a participant passed a step?
The supervisor. Two keys close a step: the AI tutor's review, which can be accepted with its gaps recorded, and the supervisor's signature, which counts over it, returns work with a reason, and closes the step.
Can we design our own method?
Yes. The supervisor sets the steps: work steps that write a part of the deliverable, decision steps that choose a direction and say why, and gates that wait for something first, such as a draft before any AI help or a supervisor's approval. Each step names what it is judged against.
What kinds of project does it run?
Capstones and theses, applied projects inside the participant's own organization, assignments between the modules of a modular programme, and custom tracks for a corporate client. Group projects are not built yet.
What can we show a corporate client at the end of a custom programme?
Each participant's finished deliverable with the record of how it was made: every draft, every review by the AI tutor and every decision by a supervisor. Reports show each participant and each cohort by step. The client's programme runs in a private space of its own, and what your school sees of that space is agreed up front: nothing, the numbers, or the full record.
How do we know the participant did the work?
You see how it was made: every draft, every review by the AI tutor and every decision by the supervisor, by name and time. The AI tutor never writes a draft; at the end it can smooth the joins between sections, with each change shown to the participant for approval. Ulern does not claim to detect text written elsewhere; it gives the supervisor the drafts, the rounds and the feedback, which is what a judgment about the work stands on.
What if we change the method while a cohort is running?
Participants keep the version they started on. A changed method is published as a new version and applies to the next cohort, so nobody's project changes under them.
Does it work for individual participants, not only cohorts?
Yes. A seat is a person, and a person can be on one programme, many or none. Cohorts are optional; the supervisor sees every participant's project either way.
What do participants hand in?
The finished deliverable, assembled from its steps and exported to Word, PDF or Markdown, together with the record of every draft, review and decision, and the supervisor's rules kept: word limits, format, a transparency note.

How Ulern compares: Riipen and Practera · Cogniti · ChatGPT Edu, Claude and Copilot

Run your next project on Ulern.

Bring the method you use today. We show you the participant's view and the supervisor's view, on your material, in 30 minutes.

90-day pilot, credited in full if you continue within 30 days · Purchase orders accepted