You-Shaped AI
← You-Shaped AI

Useful help.
Built around your life.

Choose what you need today. Every exercise includes a worked example, a quality check, and a way to improve the result.

Try a small first win

YOUR STARTING POINT

Let’s start
with your life.

Let your assistant do the interviewing. A thoughtful conversation can become a working profile, a reusable brief, and a practice plan built around your actual life.

  1. Choose one home baseOpen ChatGPT or Claude. Start a new conversation and keep the recommended settings.
  2. Have the conversationPaste the interview prompt. Answer one question at a time, by text or voice where supported. Skip anything you don’t want to share.
  3. Make it sound like youCorrect the profile. Review what can be saved, keep your own copy, and turn it into your 30-day coaching plan.
Which model should I choose?

Use a capable model you already have. In Claude, Opus is an option for a thoughtful interview, while Sonnet is a capable starting point. In ChatGPT, use the model available in your account; Astra is an option if you use Work and have access. Keep the recommended effort setting. You don’t need to buy a plan or maximize every setting to begin.

More model guidance when you need it

Before you begin: check your app’s memory and data settings. With memory enabled, details may be saved during the interview, before you approve the final profile. To keep the interview out of memory, review the app’s memory controls first. For a temporary conversation, use its temporary or incognito mode where available. A prompt alone is not a privacy control; retention may still apply.

“Start by interviewing me.”

I want you to become a more useful assistant for the way I actually live and work. Start by interviewing me. Do not give me a generic productivity plan.

Ask one question at a time and wait for my answer. Use follow-up questions when I am vague. Start with what I want help with, then learn the relevant parts of:
- My roles, responsibilities, current priorities, and goals
- What a normal week looks like and where work gets stuck
- What energizes me, what drains me, and how I make decisions
- My strengths, recurring friction, and support needs, in my own words
- How I communicate, learn, organize information, and prefer feedback
- The tools I already use and my current comfort with AI
- The constraints you need to respect, including time, budget, privacy, and decisions I want to keep for myself

Ask only what would help you support me. I may skip any question. Do not ask for passwords, account numbers, or unnecessary sensitive information about me or anyone else. Do not diagnose me or infer traits I have not described.

After about 10 to 15 questions, tell me whether you know enough or whether one or two important gaps remain. Then draft a Personal Working Profile in plain language with these sections:
1. What matters to me right now
2. My responsibilities and current projects
3. How I work and make decisions
4. What gives me energy and what drains it
5. How to communicate with and support me
6. Useful recurring tasks you could help with
7. Boundaries and things that need my approval
8. Open questions and facts you are not sure about

Keep facts, my stated preferences, and your tentative interpretations clearly separate. Skip praise and personality-label filler. Ask me what you got wrong. Revise it until I approve it.

After I approve it, ask which parts I want retained. If this app supports memory or saved instructions, explain how those work here and save only the durable context I approve through the available feature. Tell me accurately what was saved and what was not. Never claim that saving a document guarantees memory in a new chat. Also give me a concise reusable version I can keep and paste into another conversation. Exclude anything I have marked private or temporary.

Finally, use my approved profile to create a complete, personalized 30-day AI coaching prompt I can copy into a new conversation. Include the relevant approved context inside that prompt so it works on its own. The coach should ask about my daily time budget, use real tasks from my life and work, and adapt to my tools, ability, goals, and constraints. It should give me one day at a time with a practical outcome, a short explanation, an exact action or prompt to try, a quality check, a correction or feedback move, a brief reflection, and something reusable to keep. Review and adapt the plan on days 7, 14, 21, and 30. Use the five levels as a map: useful conversations; reusable context and setups; approved connections; reliable scheduled work; and bounded agent-owned jobs. Do not promise that I will reach Level 5 in a month or require extra tools that do not help. Keep my judgment, privacy, and approval boundaries explicit. The coach must not claim that a prompt creates memory, connections, schedules, or automatic reminders; those features must actually be available and verified. Where reminders are unavailable, tell me to return for the next session. Show me the complete coaching prompt for review before starting Day 1.

Before you paste: review your app’s memory settings. It may save details during the interview. Temporary or incognito modes may still retain data. Memory & privacy

Paste into a new conversation
Then open your home baseChatGPT Claude

These prompts are a starting point. Edit them freely. Download the prompt collection ↓

See how one correction makes the next answer better.

When a result misses, name the difference. Put the useful correction into your working brief or saved setup. Then test whether the next task benefits from it.

Fictional teaching example
THE FIRST DRAFT

“You dislike meetings.”

YOUR CORRECTION

“I dislike meetings that finish without a decision.”

WHAT TO KEEP

“When helping with meetings, surface the decision, owner and next step.”

Keep the corrected instruction in a note you control, or a supported instructions/memory feature you can inspect. Ask what was actually saved; don’t assume it carries into a new chat.

Your context belongs with you. This site has no interview form, accounts or analytics. You choose what to share with your AI provider.A note on memory & privacy

FIND YOUR NEXT STEP

Five levels.
Your own pace.

Think of one useful thing you keep avoiding, repeating or carrying in your head. Find the part that stalls, then choose the next level of help.

Where should I start?A quick check, based on real work

Which is the furthest step you can point to?

Choose a result you have actually seen, rather than a feature you have switched on. Use the lower step if you’re unsure.

WHAT WOULD YOU LIKE HELP WITH?Same five levels. Your kind of work.

Everyday plans, learning and personal projects. Start with something that would make your week easier.

THE CURIOUS LEVEL

Curious

You ask a chatbot questions when you think of it.

“I asked well, something real came back, and I could judge whether it was good.”

YOUR NEXT SMALL STEP

Start with the context you choose.

Have it interview you, one question at a time. Correct its working profile, keep a copy, and use it to build a personal 30-day practice plan.

Start your personal interview
EVERYDAY LIFE · LEVEL 01TRY IT ON SOMETHING REAL

Plan one thing you've been putting off

Use a short interview to make an everyday plan fit your time, energy, and preferences.

Copy the example, then open ChatGPT or Claude and paste into a new conversation.

See the exercise & worked example
About 10 minutes to make a plan; do the first small step afterward.

What you need A conversation in ChatGPT or Claude. No connections or paid feature required for the exercise.

Replace the [brackets] with your details, or use a ready-made example. Copy, open ChatGPT or Claude, and paste into a new conversation.

MAKE IT YOURS · THE REUSABLE PROMPT

Interview me about [an everyday plan or personal project], one question at a time. Learn what would make it useful, my time and budget, what I enjoy, and what usually gets in the way. Then propose a small next step and a realistic plan. Separate facts from assumptions. Let me correct it before treating it as my plan.

A WORKED EXAMPLE

Fictional teaching example. This site has not accessed an account or done the work.

BRING THIS CONTEXT

I keep putting off sorting my kitchen drawers. I have 20 minutes on Saturday, no budget for storage boxes, and I want to find the utensils I actually use. I get distracted when I empty everything at once.

A USEFUL RESULT MIGHT LOOK LIKE

Choose one drawer. Take a before photo for yourself, then spend 5 minutes removing obvious trash, 10 minutes grouping the utensils you use, and 5 minutes putting things back. Leave the other drawers alone. Stop at 20 minutes even if the whole kitchen is unfinished.

Now check it.
  1. Does it respect the 20-minute limit and zero budget?
  2. Is the first action small enough that you could start now?
  3. Can you find the utensils more easily afterward? That matters more than finishing the conversation.
If it misses, say this

That plan is too big. Keep it to one drawer, 20 minutes, and things I already own. Ask me one question if you still need to know what makes this difficult.

Don’t have the feature yet?

Use any assistant that can hold a text conversation. A paper note of your constraints is enough context; you do not need to upload a photo.

It worked when

The plan fits your actual constraints and you take one useful next step.

What you keep

A corrected plan and the preferences worth reusing.

Your part Decide what a good outcome means for your life and correct the plan until it fits.

Your ChatGPT or Claude path, and how to know it worked

In ChatGPT

Start in ChatGPT with the model available in your account. After the interview, review Personalization and the available memory controls. Keep an approved profile copy you can reuse.

In Claude

Start in Claude. Sonnet is a capable starting point; Opus is an optional choice for a thoughtful interview if you have access. Review Settings → Memory and Instructions for Claude.

A LITTLE NEW LANGUAGE

Platform = the app you use. Model = the engine doing the thinking. For now, choose a capable model you already have and leave effort at its recommended default.

How to know it worked

Try a week of ten-minute handoffs. Look for three stuck tasks that genuinely moved.

You can use different levels for different work. Keep your judgment, boundaries and approval in the picture.

01 / 05
A HANDS-ON CONTEXT LESSONChange the details. Change the answer.

YOU-SHAPED AI · A RESOURCE BY TIM WALKER

A little context changes the help.

A little context can turn a generic answer into useful help. Try it below.

Let AI interview you Skip to the resources ↓

See the difference a little context makes.

Try a detail. Watch the same task become a better fit.

THE PRACTICE TASK

Sort one kitchen drawer

No new purchases. One small area at a time.

Try another part of life

How much time do you have?

What kind of help works for you?

Practice time, not a completion guarantee.

20 MINUTES · SMALL STEPS

Sort one drawer without creating a bigger job.

  • Pick one drawer and identify the utensils you use most.
  • Work on one small area at a time: remove obvious trash and group the useful items using the space you already have.
  • Put everything back, try finding three regular items, and stop at twenty minutes.
What makes this a better fit?

Example context: No budget for organizers. · Emptying everything at once makes it harder for me to finish. · The aim is to find the utensils I use, not make the kitchen perfect.

Check: Does the drawer close, and can you find the regular items without searching?

Stop point: Stop at one drawer and twenty minutes. No new storage purchases.

Explore this area

Paste your copied prompt into ChatGPT or Claude.

A prewritten teaching demo, not a live AI chat. These are fictional practice situations. This preset demo does not send your selections to an AI service or save them.

That is a little context.
Build from the context you choose to share.

Start your personal interview

A LITTLE PRACTICE, REAL PROGRESS

Thirty days.
Built around you.

Your coach should adapt to your work, starting point and time. Here’s an example rhythm, not a fixed syllabus. There’s no requirement to reach Level 5.

Use the coaching prompt
DAYS 01–07

Get a real win.

Give useful context. Work on something stuck. Judge the result and say what needs to change.

DAYS 08–14

Keep what works.

Turn a good result into a reusable prompt, brief or checklist. Try it on the next real task.

DAYS 15–21

Build confidence.

Verify sources, compare approaches, and learn which tools genuinely help your work.

DAYS 22–30

Find your rhythm.

Refine your setup. Try a bounded handoff if useful. Review the month and choose what to keep.

THE RIGHT TOOL FOR THE JOB

One home base.
Room to grow.

Begin with ChatGPT or Claude. Learn one well. Add a new tool when there’s a real job it can help with.

START HERE

ChatGPT

A place to start your interview, think through work and develop reusable help. Use voice, media or work tools where your account supports them.

  • Choose an available, capable model
  • Review personalization and memory
  • Keep a portable working profile
Open ChatGPT Personalization guide
OR START HERE

Claude

A place to work through ideas, write, reason and build repeatable workflows. Start with a capable model; learn the finer choices as your needs grow.

  • Sonnet is a useful starting point
  • Opus is an option for deeper reasoning
  • Review memory and instructions
Open Claude Personalization guide
EXPLORE AT LEVEL 5

An agent, when ready.

Where available, a dot can help coordinate a bounded project, and a worker such as Grok Bot can tackle a defined part. These are roles you assign.

  • Check access and actual capabilities
  • Use a reviewed handoff between tools
  • Keep consequential actions behind approval
Meet dots Grok Bot guide
Ready for more? A plain-language guide to model choice

The platform is the app. The model is its reasoning engine. Effort is an optional setting in some products. Tools, instructions and permissions shape what the complete system can do. Start with defaults; compare models on work you can judge.

Haiku 5.5

Quick summaries, extraction and straightforward questions.

Sonnet 5.5

A versatile Claude default for writing, coding, analysis and workflows.

Opus 5.5

Demanding reasoning and research, refined in conversation.

Fable 5.1

Large, involved projects and sustained planning with tools.

ChatGPT’s Work and Codex guidance also distinguishes Astra, Sol and Luna. Those choices are product-specific; don’t assume that every model or effort control appears in ordinary Chat. Use what your account actually offers.

Model guidance checked October 10, 2026. Claude’s current model guide · ChatGPT’s current model guide

A small vocabulary for the whole system

You can learn these as you need them. A model is only one part of the system that helps you work.

LEVEL 2 / UNDERSTAND THE BASICS
Product or app
The place you work, with its interface, account controls and features.
Model
The AI engine interpreting requests and generating responses.
Effort
A supported setting for how much work a model spends on a response. More may take longer; it doesn’t guarantee correctness.
Prompt
Your request: the goal, constraints and what a useful result looks like.
Context
The information available for the current task, including your conversation, documents and retrieved sources.
Memory
Selected context carried into later work by a product’s memory features. It isn’t perfect recall or retraining.
LEVEL 3 / UNDERSTAND THE SETUP
Tool
A capability such as search, a calculator, file editing or an app connection.
Skill
Reusable instructions and resources for a type of task. A skill doesn’t automatically grant access.
Workflow
A repeatable sequence of inputs, actions, checks and a useful result.
Harness
The software around a model that manages the interaction, tools, task state and approval boundaries.
AI system
The whole setup: model, instructions, context, tools, permissions and human review.
Agent
A system that can take multiple steps toward a goal using its available tools and permissions.

Skills explained · The system around a model

Features, plans, usage limits and regional availability change. dots access is rolling out; Grok Bot has its own access requirements. Separate services do not automatically share memory or communicate with each other. You can always carry a reviewed brief between them yourself.

LET YOUR SETUP CARRY MORE

Give away the preparation.
Keep the judgment.

You already know how to get a useful answer. Now choose one piece of work you keep carrying and build a reliable way to hand it over. Your job is to decide what matters, notice what does not fit, and approve consequential actions.

One real project.
More reliable help.

Try one small project, such as preparing a community event. Start with one approved source. Grow the responsibility only after the previous step works.

Check access before changing your setup. dots and Grok Bot have plan, rollout, regional, or workspace requirements. You can learn the briefing and review pattern with the assistant you already have; this guide does not require buying an agent product.

03
Connected

Find one thing you would have missed.

The proof A source-linked finding you can verify, including evidence that it remains open. An empty result without verified search coverage is not an all-clear.

Walk through the setup
  1. Open the assistant you use. Ask which source connections it actually supports. Choose one relevant source and review the access before connecting it.
  2. Ask one question whose answer you already know. Open the returned source and check it. If access is missing, resolve that before assuming the assistant searched everything.
  3. Then ask it to find still-open event commitments within a specific date range. Require the original request and later replies. Read and report only.
  4. Correct one false positive or missing detail. Save that correction in the reusable instruction.
dots: connect computers and apps Grok Bot: first task and tool access

A BRIEF YOU CAN ADAPT

Using only [approved source], look for still-open commitments about [one event] between [dates]. Check later replies for completion or a changed plan. Link the original source and date. Separate commitments I accepted from requests I have not accepted. Suggest next steps for my review. Do not send messages or change records. Say exactly what you could and could not search.

If you don’t have access Without a connection, paste a redacted excerpt and practice the same checks. That develops the method; it does not prove live source access.

04
Automatic

Make the proven check arrive.

The proof The report arrives on time without you starting it, for two weeks. Its facts still match the sources and failures remain visible.

Walk through the setup
  1. Run the sweep by hand first. Remove noise and fix errors until you would genuinely want the report again.
  2. Give the exact schedule, named time zone, approved source, private destination, end date if relevant, and what to report when nothing is open or access fails.
  3. Ask the assistant to show the full saved instruction and confirm the schedule. Inspect it in the product, including its next run and pause or stop control.
  4. Compare expected and actual deliveries for two weeks. A silent failure, stale source, or wrong destination is a reason to repair the setup before relying on it.
dots: recurring tasks Grok Bot: skills and routines

A BRIEF YOU CAN ADAPT

Help me schedule the read-only event sweep we have already tested. Run [days and time] in [named time zone], using [approved source], until [end date or ongoing]. Deliver privately to [destination]. Include only [useful contents], with source links. If nothing is open, say so briefly. If access fails or data is stale, report that explicitly. No messages, record changes, or commitments. Show the complete instruction, test it safely, confirm the saved schedule and next run, and show me how to pause or stop it. If unsupported, say so rather than promising delivery.

If you don’t have access Save the prompt and run it yourself on the agreed day. Keep it labeled as a manual routine until a supported schedule is actually confirmed.

05
Autonomous Agents

Receive a finished package to review.

The proof The agreed package reaches its finish line. Across a week of suitable jobs, your work is decisions and review rather than reconstructing missing preparation.

Walk through the setup
  1. Choose a bounded outcome: prepare the complete event follow-up package. Define the required documents, scope, due time, source access, and acceptance tests.
  2. Allow the agent to gather facts, reconcile status, draft, and check within that scope. Reserve sending, publishing, spending, deletion, access changes, and commitments for your explicit decision.
  3. Require the actual drafts and files, supporting sources, checks performed, unresolved issues, and exact decisions for you. A plan for work is not the completed work.
  4. Open the deliverables, verify important facts and recipients, and approve, revise, or reject each proposed action. Put recurring corrections back into the next brief.
dots: review work and stop controls Grok Bot: approval and privacy controls

A BRIEF YOU CAN ADAPT

Prepare the complete follow-up package for [one event] by [date, time, and time zone]. Use only [approved notes and relevant messages]. Done means the current commitments are reconciled, the necessary drafts are written for verified recipients, required documents are ready, and every important fact has a source. You may read, research, draft, and check. Ask before sending, publishing, buying, deleting, changing access, or making commitments. If blocked, explain what is missing and continue independent work within scope. Return the actual package, source links, tests performed, uncertainty, and decisions for me. Do not call a plan or incomplete batch finished.

If you don’t have access A normal assistant conversation can help prepare the same package while you manage the steps. Use the agent milestone only for work that actually completes end to end in your setup.

Where to inspect schedules, work and controls

If you use a dot

Ask your dot to create the recurring task, then open Scheduled to review its instruction, timing, and destination. For a delegated job in the desktop app, open the dot's profile → Activity → the task to inspect files, progress, or a request for input. Canceling a schedule and stopping an active task are separate actions.

Current setup and controls

If you use Grok Bot

Ask the Bot that should own the job to create its routine. Check the next run. Open that Bot → View conversation details → Routines to inspect or pause it. Use Test run with safe inputs: a test can perform real actions. Keep this first routine read-only and review its recent run history.

Current setup and controls

Official setup and control pages checked October 10, 2026. Interfaces and availability can change; follow the linked current guides.

AN OPTIONAL NEXT PATTERNOne coordinator. One focused worker. You decide.

A dot and an optional worker such as Grok Bot can fill these roles where available. This is a pattern you assign, not a product ranking or a promise they communicate directly. One capable assistant can be enough.

  1. In your main assistant, explain the outcome, priorities, and boundaries. Ask it to write a self-contained worker brief with all necessary approved context.
  2. Read the brief. Remove unnecessary personal details and confirm the scope. Copy it into the worker yourself; do not assume another service knows your history.
  3. Let the worker complete its bounded part. Ask for the actual artifact, source links, tests, gaps, and decisions. Keep consequential actions behind review.
  4. Bring that result back to the main assistant. Ask it to reconcile the work against the original brief and identify missing evidence or unfinished pieces.
  5. Make the judgment calls. Approve only the specific actions you intend, correct the brief, and keep the parts that made the next job easier.

REVIEW THE WORKER’S RESULT

Here are the original brief and the worker's result. Check them against each other. Which acceptance tests are actually supported by evidence? What is missing, uncertain, or outside the approved scope? Give me the decisions that require my judgment. Do not assume an action happened because the worker described a plan. Do not contact the worker or take external action unless I explicitly authorize it.

Different services do not automatically share context or memory. Within Grok Bot, separate Bots share computer files and signed-in sessions; they are not separate security boundaries. Read the current security guidance

FOR DESIGN & DEVELOPMENT

Your taste is
part of the brief.

A slide, a poster, a website or a little app should feel intentional and work for the person using it. Bring a point of view. Then make the result earn it.

01

Collect references. Name what works.

Choose two or three examples you like. Be specific: type, spacing, hierarchy, tone, navigation. Borrow the principles, not someone else’s identity or artwork.

02

Write the intention. Choose a direction.

Who is it for? What should they understand or do? Ask for two genuinely different approaches, then choose one for a reason.

03

Critique. Refine in focused passes.

Find the three biggest problems. Fix structure, then type and spacing, then detail. Keep what works instead of regenerating everything.

04

Use it where it has to work.

Open the exported file. Check the real size. For an app, try the full flow on a phone and with a keyboard. Test empty, invalid and long inputs.

A design brief with taste
Help me create [artifact] for [audience]. The main thing they should understand or do is [outcome]. It will be used in [format, device, or setting].

Here are my references: [links or examples I can share]. What I like about each is [specific qualities]. Use these to understand my taste; don't copy their artwork, branding, or exact layout.

Required content: [content]. Constraints: [size, accessibility, tools, timing]. Things to avoid: [generic visual habits or styles that don't fit].

Ask the few questions that would materially improve the brief. Then propose two genuinely different visual directions. Explain the hierarchy, typography, spacing, color, and interaction choices in plain language. Recommend one for this audience and purpose.

After I choose, make a first draft. Critique it against the brief and references before presenting it. Show me the three biggest improvements you would make next. Refine in focused passes using my feedback, preserving the parts that work. Verify the final file or experience in its intended format and tell me what you actually checked.
Build a small, useful app
Help me build a small tool for [person] who needs to [one specific job]. I am describing the goal in everyday language; explain technical choices simply and don't assume I am a developer.

The essential flow is: [starting point] → [main action] → [useful result]. Start with sample data. Keep the first version small enough to test. Ask essential questions before building and show me a brief plan.

For visual direction, use these references: [links or examples]. I like [specific hierarchy, typography, spacing, navigation, or interaction qualities]. Translate those qualities into an original design with a clear main action. Avoid decorative filler and stock layouts that don't fit the purpose.

Before coding, define what must work and how we will test it. Include the normal flow, empty input, invalid input, long content, and a phone-sized screen. Don't hide unfinished features behind buttons that appear to work.

Build the smallest useful version, then test the entire flow. Show me what works, what remains incomplete, and any assumptions. Critique the visual quality as well as the behavior: hierarchy, type, spacing, contrast, consistency, and ease of use. Fix the three highest-impact issues before adding features.

Use mock data until I approve any real data connection. Ask before adding costs, accounts, permissions, real personal data, or public deployment. Keep secrets out of the code and the browser. Return a working preview or runnable files, clear usage instructions, and the tests you actually performed.
Your quality check, before you call it done

Can you explain the visual choices? Is the main action clear? Does every section earn its place? Is the copy specific? Are visuals relevant? Is it readable at the intended size? Do the links, controls and exports actually work?

A compelling screenshot is one check. A useful, usable result is the finish line.

THE PRINCIPLE UNDERNEATH IT ALL

Use AI without
removing yourself.

AI can research, challenge, organize, draft, design and build with you. Your values, attention, taste and judgment shape what the work becomes.

Try this: “Keep me in the work” 
Help me with [task] in a way that strengthens my contribution. Start by asking for my purpose, current thinking, standards, and anything that requires my judgment or personal presence.

Then suggest which parts you can research, organize, draft, design, test, or prepare. Make the decisions that remain mine explicit. Don't invent beliefs, experiences, feelings, promises, or conclusions on my behalf. Challenge weak assumptions and show meaningful alternatives.

Work from the relevant source material. Separate facts from inferences. Give me a result I can inspect and improve, with the important uncertainties and checks made visible. Help me understand enough to stand behind what I use. Before any consequential action, confirm the scope and approval it needs.
01

Bring your judgment.

Think with it. Ask for questions, hints or critique when learning is the goal. Check the evidence and stay involved enough to understand and stand behind the answer.

02

Keep your presence.

Preparation can help you show up well. Your care, listening, convictions and responsibility still belong in the room.

03

Own the outcome.

Set the boundaries. Review consequential actions. Keep the approval for commitments, public work, spending and decisions that affect other people.

A RESOURCE BY TIM WALKER

More useful. More personal.

This field guide grows out of Tim Walker’s AI for Your Genius teaching and its five levels of practical use. The method is hands-on: let AI interview you, correct the context, keep useful setups, and earn more capable help through results you can verify.

“Digital tools should buy me analog time.”

Tim Walker

A FEW GOOD QUESTIONS

Keep the human
in the picture.

Does getting AI help make the work less mine?

Getting help can strengthen your contribution. Ask what role the help plays. Are you bringing the purpose and perspective, checking the claims, developing your understanding and taking responsibility? Use AI to question, research, organize, draft and refine with you. Be honest about its contribution when the audience or the rules of the work call for it.

What if I don’t trust AI yet?

Start with a task you know how to judge and material you can check. Ask a few questions in a field you understand. Mark what is right, wrong or missing. Keep the useful parts and the boundaries you need. Trust should grow from evidence for a particular job; you don’t need to decide that AI is trustworthy at everything.

Am I training my own AI model?

You’re personalizing an assistant with useful context, instructions and available memory. That’s different from training or fine-tuning the underlying model. A reviewed working profile is a portable way to carry your context into another conversation.

What should I save in memory?

Save durable preferences and useful working context that you’re comfortable retaining. Review, correct and remove saved information in the app itself. A saved profile file doesn’t guarantee recall in a new chat. Memory and model-improvement data settings are different controls. Read the current ChatGPT memory guide or Claude memory guide.

Do I need a paid plan or several tools?

Start with an assistant you already have access to. These learning prompts don’t require a purchase. Some models, connections, schedules and agent tools have plan or access requirements. Check those before adding anything, and use a manual workflow when that’s enough.

Will the coach contact me every day?

The coaching prompt creates a practice plan inside your AI conversation. It does not schedule reminders. Return to the conversation for the next session, or separately configure a supported reminder and verify that it is saved.

Does this website collect my answers?

No interview answers are entered or stored on this website. Copy a prompt and use it in the AI service you choose. This site has no analytics, sign-up form or account system. The hosting provider may process standard request logs; the AI service you choose has its own data, memory and retention policies.

Is this affiliated with the AI providers?

No. You-Shaped AI is an independent educational resource. Product names belong to their respective owners. Examples describe possible workflows, not certifications, guaranteed results or endorsements.

YOUR NEXT STEP CAN BE SMALL

Start with your life.
Leave with something useful.

A clearer plan. A working profile that sounds like you. One less thing stuck in your head. Choose a starting point and try it on something real.

Let AI interview you Find a small first win One assistant. Your own pace.

Copy this text

Your browser couldn’t copy automatically. Select the text below and copy it.