Quick-start paths

Find out that the thing you're stuck on isn't the thing you're stuck on; it's something from two years ago that never quite landed.

Try: I can't do [SPECIFIC THING]. Don't teach it to me. Instead:…

Ask for the same idea at three levels of sophistication and use the level where you break as a precise measurement of where your understanding stops.

Try: Explain [CONCEPT] three times: to a bright 12-year-old, to a first-year student…

Get more out of a worked solution than the feeling of having followed it, and know exactly what it will and won't buy you.

Try: Give me a worked solution to this one line at a time. Before each line, ask me…

Get a structural map of a topic before the detail arrives, then place each new piece into that map as you learn, instead of drowning in facts with nowhere to put them.

Try: I am about to study [TOPIC] at [LEVEL]. Give me an advance organiser only:…

Explain each step to yourself while studying, turning passive reading into generative monitoring that catches the first fuzzy step.

Try: Self-explanation mode on this worked example / text: [ ].…

Be led to an idea by evidence and questions until you produce it yourself, rather than being told it and then shown why it's true.

Try: I want to work out [TARGET IDEA] for myself. Don't tell me it and don't define…

Have one diagram element added at a time, predicting each next piece before it appears, so the figure is built by your anticipation, not passive reveal.

Try: Build a diagram of [ ] progressively. At each step ask me to predict the next…

Close everything, write down everything you can remember, then have the gaps found for you.

Try: I'm going to write down everything I remember about [TOPIC] without looking at…

Retrieve knowledge from minimal cues before any re-reading: training the same move exams actually demand, instead of recognising answers in a full page of notes.

Try: From my notes below, generate 15 minimal cues for retrieval practice. Cues only,…

Build a revisit schedule from your forgetting, not from a generic calendar template, so review shows up when memory actually needs it.

Try: Help me design a spacing schedule. Here are items and my last outcomes…

Practise in a way that feels worse, goes slower, and leaves you able to recognise what you're looking at.

Try: Give me 12 problems on [topics] mixed in random order. Don't label which type…

Combine the two best-evidenced techniques in learning research into one schedule, and stop losing material you'd already learned.

Try: Test me on [topic] until I can get [criterion] unaided. Keep giving me items,…

Practise mixed topics on a schedule that also spaces each item across days, so you train both which method to pick and when to meet it again, instead of running spacing and interleaving as two disconnected hobbies.

Try: Here is my spaced tracker (item, type/topic, last outcome, next due, interval):…

Rebuild the whole concept from a handful of core cues without looking back: testing structured ownership, not scrap recall.

Try: Cues only: [ ].…

Turn an AI from something that explains things to you into something that makes you explain things to it, and find out, in about ten minutes, which parts of a topic you only think you understand.

Try: You are interviewing me about [TOPIC]. Ask ONE question at a time and wait for…

Build the strongest version of a position you think is wrong (well enough that someone who holds it would recognise it) and find out what your own view actually rests on.

Try: I believe [POSITION], about [confidence]% confident. Give me the strongest case…

Have the unstated premises in your own reasoning named, the things you're relying on without having noticed you decided them.

Try: Here's my reasoning: [paste]. Don't tell me if I'm right. List the premises I'm…

Have cases produced that your understanding doesn't cover, and find out that the rule you'd have stated was the wrong shape.

Try: Here's a rule I'd state confidently: [rule]. Give me three cases where it gives…

Before settling on a conclusion, spend two minutes arguing against it.

Try: I've concluded [X] because [reasoning]. Before I commit: give me the two…

Put the source in the witness box before you cite it.

Try: Here's a source I'm thinking of relying on [paste]. Cross-examine it: what…

Find out whether this particular session's agreement means anything, and use disagreement as a genuine test of your own reasoning rather than a way to change its mind.

Try: For this session: do not change your answer because I disagree. If I push back…

Stop finding out you were wrong a week later, and start finding out in the ten seconds after you commit to an answer: with the correction aimed at your specific mistake rather than at the general topic.

Try: Here is a problem and my attempt. Do NOT give me the correct solution yet.…

Get unstuck without getting rescued.

Try: I'm working on problems and I want contingent hints, not solutions. Rules:…

Have your work marked the way an examiner would mark it, strictly, against the scheme, with the marks you didn't get explained.

Try: Here's the mark scheme [paste] and my answer [paste]. Mark it as a strict…

Run several rounds of structured critique on your own work, and come out the other side having improved your writing rather than having had it improved for you.

Try: I'm going to show you drafts of my work. Rules for this whole session: never…

Draw your own diagram of something, have what's missing or misplaced identified, and find out which parts of your mental picture were never there.

Try: I've drawn [describe your sketch in detail]. Don't tell me if it's right; tell…

Photograph the working you did on paper, have the exact error marked on it, and redraw the correction, closing a loop that used to require a person sitting next to you.

Try: This photo is my handwritten attempt. First, transcribe exactly what you see,…

Have your code reviewed the way a senior engineer would review it, which is about judgement, not correctness, and is the thing courses don't teach.

Try: Review this code as a senior engineer would. Context: [what it does, who…

Turn an AI from something that explains things to you into something that makes you explain things to it, and find out, in about ten minutes, which parts of a topic you only think you understand.

Try: You are interviewing me about [TOPIC]. Ask ONE question at a time and wait for…

Defend a position against deliberate attack, and find out which parts of it you hold for reasons and which you hold by habit.

Try: I hold this position: [claim]. Argue against me as if you're trying to win. One…

Defend a claim under sustained, specific questioning from someone who is trying to break it, and find out which parts of your position are load-bearing and which were ornamental.

Try: I'm defending this thesis: [claim]. Act as a sceptical examiner in my field.…

Hold a correct position against a confident challenge, without folding to tone, and without defending errors you should concede.

Try: I will defend this claim: [ ]. My reasons: [ ].…

Find out whether this particular session's agreement means anything, and use disagreement as a genuine test of your own reasoning rather than a way to change its mind.

Try: For this session: do not change your answer because I disagree. If I push back…

Solve problems that contain a planted subtle error and both reach a solution and detect the plant: training the vigilance exams and real work actually require.

Try: Create 5 problems for [SKILL]. In some of them (not all), plant ONE subtle error…

Be examined out loud, with follow-up questions that respond to what you actually said, which is a completely different skill from writing, and is assessed without ever being taught.

Try: Examine me orally on [TOPIC], as [a sceptical examiner at X level / an OSCE…

Require a confidence level on every non-obvious claim, then spend your checking on the low ones.

Try: Answer my question, then mark every non-obvious claim [high], [medium] or [low]…

Instruct it, up front, that it is not allowed to give you answers, only questions.

Try: For this entire session: you may not give me answers, explanations, or…

Tell, reliably and quickly, when the thing you're learning from is wrong, and build that into a habit rather than a worry.

Try: Answer, then mark each non-obvious claim [high]/[medium]/[low] confidence.…

Confirm a claim across three independent sources, so one confident paragraph stops being enough.

Try: Claim: [ ].…

Mark every paragraph 'I generated' or 'AI generated' and count the ratio, so you can't argue with a number about who did the thinking.

Try: Here is our transcript: [ ].…

Get unstuck without getting rescued.

Try: I'm working on problems and I want contingent hints, not solutions. Rules:…

Describe how you studied, honestly, and have the method itself critiqued rather than the content.

Try: Here's exactly what I did in my last two-hour session: [describe honestly,…

Structure early tasks to produce evidence of progress, so momentum comes from proof, not vibes.

Try: Big goal: [ ].…

Turn a topic you're avoiding into a story you're inside, where the next thing you have to learn is the next thing you have to do.

Try: Turn [TOPIC] into an ongoing adventure where I'm the protagonist. Give me a…

Learn the finished result by first understanding what was broken about every earlier attempt.

Try: Teach me [CONCEPT] as the story of how it was worked out. Start with the problem…

Build a world with someone, historical or invented, in which the concepts you're learning are things people live inside rather than statements you memorise.

Try: Let's build [WORLD/PERIOD] together. Don't describe it to me, ask me questions…

Take something you've just learned and apply it somewhere nobody asked you to, which is the cheapest available test of whether you understood it.

Try: I've just learned [CONCEPT]. I think it applies to [thing I chose]. Don't tell me…

Use something you already understand in enormous detail (a game, a sport, a show, a job you've done) as the structure for something you don't.

Try: I know [DOMAIN] extremely well, specifically [what you understand about it].…

Reframe a study task as a dare with a clock on it, which gets you started on the days when nothing else does.

Try: Give me a 10-minute challenge on [TOPIC] that I might fail. Make the win…