How Aynstyn judges what you really know

Picture two people explaining the same idea. The first writes three confident paragraphs, full of the right words. The second writes four sentences, plain and precise, the way someone who has actually done the work talks.

Most assessments reward the first. Length looks like effort, and vocabulary looks like knowledge. That is how the competence illusion survives: you feel ready because you can say the words, and nothing in front of you tells you otherwise.

Confidence is free. Proof isn't.

Aynstyn exists to close the distance between feeling ready and being ready. So we went back to the heart of the system, the moment it judges an answer, and rebuilt it around one rule: judge what your answer shows, and show you the proof.

One number that means something

Every result now leads with a single figure: your Competence Index. It is not a percentage of facts recalled. It reads how close your understanding is to a clear, joined-up whole: how much you understand, how well the ideas connect, whether you can use them, and how much stands in the way.

It sits on one scale, from Early Foundations to Coherent Mastery, and that same scale carries your journey. Every attempt, every check and every practice session appears as a point on it, so you can see exactly how far each step moved you. When you improve after practice, the number you see is where you stand now, not where you started.

What you show, not how long you write

A short, precise answer that uses the ideas correctly now earns what it shows. A long answer that strings terms together does not. Naming a concept is not the same as understanding it, so we look at how each idea is used: to justify a choice, to explain a trade-off, to reason about a failure.

An answer still has to say enough to judge. A single phrase cannot show depth. Beyond that, length is no longer the measure. Understanding is.

Wrong is not the same as not yet shown

Missing something and getting something wrong are different problems, with different fixes. Aynstyn now tells them apart.

• Worth correcting means your answer made a claim that does not hold. We show you your own words, say why the idea does not hold, and give you the one to keep.
• Not shown yet means your answer did not cover it. It is a gap to fill, not a mistake to unlearn.

We never call something wrong without quoting what you actually wrote. If the system cannot point to your sentence, it treats the topic as not yet shown. Being told you are wrong should come with proof, every time.

And "you sound very sure" is not the same as "you believe something false". Aynstyn names a misconception only when it can name the specific idea. Confidence on its own is a note about calibration, never a verdict.

One first fix, everywhere

A list of ten gaps is not direction. So every gap now sits in one order: wrong ideas first, because new learning gets built on top of them, then what your answer did not show, then what it covered only thinly.

That order is the same on your result, in your practice sessions, in the 2-minute check, on your dashboard and in the agent. Wherever you are, Aynstyn says the same thing: fix this first.

The same answer, the same verdict

A judgment you cannot trust is not worth acting on. The grader now runs the same way every time, with one standard for every channel you use, whether you write, speak, or come through the agent. It is anchored to what you were actually taught: a course module is judged against that module's own material, and a retake is judged against the same topics as your last attempt, so progress is measured on the same ground.

And a gap only closes on evidence. A later answer that simply does not mention a topic does not count as fixing it. You close a gap by showing it.

Before we grade it: how well do you know it?

Here is the part we are most excited about. Before your answer is graded, Aynstyn now asks one optional question: how well do you know this? Just starting, getting there, solid, strong, or could teach it.

Then it shows you how your read compares with what your answer showed. Sometimes you read your level exactly. Sometimes the gaps below are the ones you did not see yet. And often, especially for people who know their field, the answer shows more than you gave yourself credit for. That is the move from "I can't" to "I can", with proof.

Knowing what you know is a skill in its own right, and it can be trained. The classic research on self-assessment found that building real skill is exactly what sharpens it. So we built the measure and the training into the same loop.

A loop that checks the gap actually closed

A diagnosis is only the start. Every gap now flows into a short, honest loop:

1. Confirm it. A free 30-second check on your top "not shown" gap: is it a real gap, or did you just not mention it? Its explanation waits until you answer, so the check is honest.
2. Close it. A 2-minute check or a practice session on that gap, graded on that gap. Tell us how sure you are first: if you were sure and still wrong, that is the one to fix first.
3. Sharpen it. If something is still missing, the next check asks for exactly that, not the same question again.
4. Keep it. A week after you fix a gap, Aynstyn brings it back: check it held. Fixed twice, it is yours.

Back to the two answers

The confident paragraphs and the four precise sentences now get judged for what they are. The first learns which words it has not earned yet. The second gets the credit it deserves, and its next gap, quoted, explained and ready to close.

That is what this rebuild is for: clarity about where you stand, a direction you can act on today, and momentum you can see on one scale. Take an assessment on something that matters to you, and let it show you the first fix.