Turn what you just asked into MCAT-style practice

Curiosity is where retention starts

You remember what you wanted to know. A topic you chased down because it bothered you sticks in a way that the next item on a review list does not, and every study method worth the name is built on that difference.

Most tools throw it away. Reading happens in one place and questions happen in another, and moving between them costs enough friction that you do it later, or never. By the time you sit down to practice, the confusion that made the topic memorable has faded and the topic list came from somewhere else entirely.

The best moment to practice a topic is the moment you stopped being confused by it.

What the button does

Once a conversation has started, a button reading Practice this topic sits at the top of the chat; on a phone it is the target icon in the top bar. Tap it and a panel opens saying Reading your conversation. Here is what that means, step by step.

• It reads the conversation, not the last message. The whole exchange goes in — up to the last thirty messages, capped at 800 characters each. Ask why tertiary substrates favor SN1 over SN2, get sidetracked into carbocation stability, come back to the solvent: all of it counts.
• One bounded call maps it onto the bank. A small utility model reads the transcript and matches it against the subjects and topics the question bank already has. One call, nothing recursive, no browsing.
• It returns topics and a reason. Up to three subjects, up to three topics, and a single sentence explaining why it chose them, shown in a tip box so you can disagree with it before anything is created.
• Then you are in the ordinary picker. The same subject-and-topic picker Exam Prep uses, with those choices already ticked.

A conversation about finding a buffer’s pH with Henderson–Hasselbalch might come back as General Chemistry, Biochemistry, or both. Whichever it picks, the sentence in the tip box says why, in plain words.

The model names the topic. It never picks a question.

This is the part that decides whether the feature is trustworthy, so it is worth being precise about it.

The model does not return questions. It returns slugs — short identifiers for subjects and topics that already exist in the bank. Those slugs are matched on the server against the real list, and any the server does not recognize are dropped. Not corrected, not guessed at, dropped. The model cannot invent a topic the bank does not have, because an invented one matches nothing and disappears.

Question selection happens after that, and no model is involved in it at all. Once the subjects and topics are fixed, choosing which questions you actually see is plain policy over the filtered pool: inspectable, and the same rules every time.

The model may say what you were talking about. It may not decide what you practice.

The reasoning is the same as the one behind answers that cite the page they came from: give the model the narrow job it is good at, and let checkable machinery handle everything downstream of it.

You get the last word on the session

Pre-selected is not the same as chosen for you. Every subject and topic arrives ticked and editable. Drop one you did not mean, add Organic Chemistry to a Biochemistry conversation that drifted there, or widen the mix if you want it harder.

• An optional question goal. Set a number if you are working to a target, or leave it open.
• An optional timer. Sixty, ninety or 120 seconds per question, for when the thing you need to practice is pace rather than recall. Timed questions skip the AI analysis of your answer and the follow-up chat; you still get the correct answer and the explanation.
• A session tied to its conversation. The session is created linked back to the chat it came from, so the questions and the discussion that prompted them stay connected.

Then tap Start prep session and the first question opens.

How the questions are served

The button starts a Practice session: questions drawn at random from the subjects and topics you kept. Within a session, a question is never served to you twice; when the pool runs out, the session ends rather than repeating itself.

Passage-based sets stay whole. Several questions on one passage arrive in order, one after another, and a question goal never cuts a set in half.

The other ways to practice the same subjects live in Exam Prep, and each changes which pool the questions come from:

• Unseen questions. Only questions you have never attempted in any session.
• Review mistakes. Questions whose latest attempt you got wrong, oldest first, so the ones you have avoided longest come back first.
• Bookmarked. Only the questions you starred.

A mode with nothing to serve is grayed out and tells you why, rather than starting an empty session and leaving you to work out what went wrong.

What this does not do

It does not write questions. Everything you get is an MCAT-style question that already existed in the bank before you asked anything. The feature chooses where to look; it does not generate the material.

It does not read anything beyond the conversation you are in, and it does not pretend to place a conversation it could not. Topics it cannot match are simply dropped, which means the honest failure here is a session narrower than you hoped, never a session about something you never mentioned. That trade is deliberate, and it is the same one described in our note on trusting AI in pre-med study.

The bank covers the science of the MCAT: Biology, General Chemistry, Organic Chemistry, Biochemistry, Physics, and Psychology & Sociology. It does not cover CARS. And practicing a topic is not the same as being right about it: Med Guru Pre-Med is a study aid for your science courses and the MCAT, an AI can be wrong, and the citation is how you check.