Future of Work Lab
← COURSES/ PERSONAL AI OS / SESSION 01
0001020304050607
25% · 2 of 8
Building Session 01
HOW TO THINK & WORK WITH CLAUDE · AI for Non-Tech Students Accelerator
Build your personal AI operating system for your thesis work

Prompt, Context & Thinking with AI

Claude is only as good as the task, context, and standards you give it. This session teaches you how to work with Claude in a more deliberate way so it becomes a real thinking partner for your thesis, not just a chatbot that gives polished but shallow answers.

Learning to work with AI is asking better questions, giving it context, setting standards, and staying responsible for the result.

For years, doing thesis work meant you alone in front of a blank page. You read, you researched, you wrote, you got stuck — and the tools around you (Word, Google, Consensus) could format and store, but they could not think with you. When the argument wouldn’t hold or the structure wouldn’t come, there was no one in the room.

Then we got AI. Most people open their AI and ask a vague question, get a vague answer, and assume that is all the tool can do. It is not. Claude is not another search box. It is a thinking partner you manage — capable of comparing papers, pressure-testing an argument, structuring a chapter, or rewriting a paragraph at a level most students never had access to. But it only performs that well when you treat it like labor: give it a role, a task, the context it needs, the standard you’re holding it to, and the boundaries it should respect. A vague prompt gets a vague answer. An intentional prompt gets real work back.

Claude is powerful and Claude is not magic. It can think, structure, compare, challenge, and draft alongside you. It can also sound confident while being wrong. By the end of this session you will know how to brief it properly, spot weak output before it shapes your thinking, and think with AI in a way that strengthens your judgment instead of replacing it.

KEY INSIGHT
Claude does not respond to what you meant. It responds to what you actually gave it. The clearer your prompt, context, and standards, the better the work it can produce.

For decades, your tools could format your thinking.
Now one of them can actually think with you.

BEFORE · 2023 TO 2025 · Traditional software

Software helped you write. It could not help you think.

Word processors, search engines, and reference managers each had a fixed job. You clicked buttons, picked from menus, and worked inside the limits of the tool. They could format your draft, store your sources, and check your spelling — but when you were stuck on your research question, your literature review, or the structure of a chapter, none of them could move the thinking forward.

The software helped you write. It did not help you think.
NOW · 2026 · Claude

Claude responds to language, not a button.

Claude responds to plain language and to the context you provide. No tech skills necessary. That means you can ask it to help you clarify a research question, compare papers, identify gaps in an argument, structure a chapter, or improve a draft. But the quality of the result depends on the quality of both the prompt and the context you give it. Working with Claude helps you think more clearly, work more systematically, and produce stronger drafts faster.

Claude helps you think so you can write better.
“A prompt is the instruction. Context is the background Claude needs to do the job. Better prompt, better context — better work back from your AI labor.

Start Here: Key Terms for Working with Claude

Prompt

The message or instruction you type into Claude to give it a task or ask it a question. A prompt is how you tell Claude what you want it to do.

Output

What Claude gives back to you. This could be an answer, a summary, a paragraph, a list, or a draft.

Context curation

Choosing the right information, documents etc. to give Claude. This means not giving it everything, but giving it the most relevant material for the task. For thesis work, context could include your topic, research question, stage, sources, or supervisor feedback.

Evidence standard

The level of proof or support you expect in the answer. For example, you may want Claude to only make claims that are backed by sources.

Boundaries

The limits you set for Claude. For example, you might tell it not to invent citations, not to make unsupported claims, or not to write in an overly casual tone.

Judgment

Your ability to decide what is accurate, useful, relevant, and strong. Claude can help, but your judgment is still essential.

Hallucinate

When AI makes something up and presents it as if it were true. This can happen with facts, citations, quotes, or sources.

Prompting, Thinking & Context Checklist — Step-by-Step Instructions

Click the bright cyan blue arrow at the end of each box to get your step-by-step instructions.

Work through each item. Check them off as you go.

0 / 4 COMPLETED
STEP 1 · Understand the 9 Components
~20 min
Learn the 9 things a prompt can include — and pick the 3 or 4 your task needs
These are the nine ingredients a prompt can have. You almost never need all of them. Most strong prompts use three or four — chosen for the task in front of you.
Map the 9 components to 3 real university tasks
Practice writing better prompts using the framework. Turn 3 real university tasks into structured prompts.
STEP 2 · Judge and Save
~15 min
Run the 3-part discernment check on Claude's response
Getting a response is not the finish line. Using a response is. This check gives you a consistent way to decide: ready to use, or one more pass?
Save your best prompt as a reusable template
A prompt you write once and refine is worth more than ten you throw away. Leave today with at least one prompt you can reuse next week.

What Else Can You Do with Prompts

  • Drafting Your Research Proposal
  • Pressure-Testing Your Thesis Argument
  • Writing Conference Abstract Drafts
  • Generating Advisor Meeting Agendas

Context Is the Most Important Skill in Your Thesis OS

When you walk into your supervisor's office…

…do you launch straight into your question? Or do you first remind them which chapter you're on, what you discussed last time, and what you've tried already?

Obviously the second. Because you know that without that background, their answer will be generic. With it, their answer will be precise, targeted, and actually useful.

That's context. And the same principle applies — at a much larger scale — to working with AI.

What context actually is

Context is everything the AI can see when it generates a response. Not just your question. Everything:

  • Your instructions and what you've uploaded
  • The conversation history
  • Any documents on your project
  • Files, notes, examples you've attached
  • Your university's thesis instructions
  • Hard data & statistics on your topic
  • Your methodology, your theoretical framework, your supervisor's feedback

Your actual question — the thing you typed — might represent 1–2% of what the model processes. The rest is context.That ratio is what most people don't understand, and it's why two people can ask Claude the same question and get completely different quality answers.

Every time you start a new chat without carrying that context forward, you're essentially hiring a brilliant research assistant and then refusing to brief them. They'll do something. It won't be what you needed.

The students who get mediocre outputs from AI are usually not asking bad questions. They're providing thin context.

The two problems: noise and gaps

There are two ways context fails, and both matter.

Gaps— missing information. The AI doesn't know your research question, your methodology, your argument so far. It fills the gap with generic content. The output sounds fine but isn't yours.

Noise— irrelevant information. Old conversation turns that are no longer relevant, uploaded documents that don't apply to this specific task, background detail that dilutes the signal. Noise doesn't just waste space — it actively pulls attention away from what matters.

This is why context engineering has two sides: curation (getting the right things in) and pruning (getting the wrong things out).

Context curation: what to put in

Before any serious AI-assisted task on your thesis, provide:

  • Your research identity — what your thesis is about, at the level of argument, not just topic. Not “I'm writing about climate policy” but “I'm arguing that national sovereignty concerns systematically undermine multilateral climate commitments, using the EU and ASEAN as contrasting cases.”
  • Where you are in the work— which chapter; what you've established; what's still open.
  • The specific task right now — not just “help me write” but “help me strengthen the transition between my theoretical framework and my case study methodology.”
  • What good looks like for you— your supervisor's voice? A specific journal's style? A level of formality? If you don't define quality, the AI guesses.
  • What's off limits— arguments you've already rejected and why. Framings that don't fit your discipline. Sources you've been told to avoid.

The Lütke test: “Could someone read only what I've provided and complete this task accurately, without asking me anything else?” If the answer is no, you haven't curated enough.

Context pruning: what to take out

As a project grows — especially a thesis that runs for months — context accumulates. Early drafts. Abandoned arguments. Feedback that's been incorporated. Old versions of sections. Keeping all of it in your project creates noise. It competes with the current, relevant information for attention.

Practical pruning principles for thesis work:

  • One task, one focused window.Don't dump your entire thesis into every conversation. For today's task — strengthening a literature review paragraph — provide the paragraph, your research question, and your quality standards. That's probably enough.
  • Archive, don't delete. Move old drafts and resolved feedback to a separate folder. Keep your active project context lean and current.
  • Update your standing brief. If you keep a “Thesis OS” document, it should reflect where you are now, not where you started six months ago.
  • The signal test.For every piece of context you're about to include, ask: “Does removing this risk making the AI's answer worse?” If not, cut it.

Curating Context Checklist — Step-by-Step Instructions

Click the bright cyan blue arrow at the end of each box to get your step-by-step instructions.

Work through each item. Check them off as you go.

0 / 2 COMPLETED
STEP 1 · Context Pruning
~20 min
The Context Audit — a 4-step pruning exercise
Pull your current thesis context out, signal-test each piece, and rewrite a lean Thesis Brief you can reuse every month.
STEP 2 · Finding the Context Gaps
~25 min
Use Claude to diagnose what’s missing from your brief
Turn Claude into an interviewer. Claude pauses, asks targeted clarifying questions via interactive popups, and only proceeds once it has what it needs. The popup responses become your gap map.
KEY INSIGHT
You are managing an information environment. The quality of what the AI returns is almost entirely determined by the quality of what you give it. Your job isn’t to only ask better questions — it’s to also give it just the right information.

Thinking With Claude, Not Claude Thinking for You

I want you to be honest with yourself for a moment. How many times have you asked Claude a question, read the answer, and thought — that’s actually better than what I was going to say. So you just used it. You didn’t push back. You didn’t add to it. You just… took it.

That feeling — that “Claude is smarter than me” feeling — is the most dangerous moment in your entire relationship with AI. Not because Claude gave you a bad answer. But because it gave you a good enough answer that you stopped thinking.

That’s what we’re fixing today. You cannot allow this to happen when writing your thesis. You have to come up with your own arguments and defend them.

What Cognitive Offloading Actually Means

Your thesis is, in the end, a defended argument. A committee will read it and ask whether you thought this through. That's the part you cannot let Claude do for you.

Everything else? Fair game. Cognitive offloading just means moving thinking out of your head into a tool — like writing a to-do list, using a calculator, or sketching a diagram. You do it constantly. It is smart, not lazy.

The trap is selective offloading: using Claude to do the structural and editorial work (good) and drifting into letting it form your argument for you (bad). Master's-level thesis work has three kinds of thinking. Knowing which is which is the whole skill.

Offload to Claude
  • Structure and formatting decisions
  • Finding gaps in your literature
  • Generating options you'll then evaluate
  • Checking your logic for internal consistency
  • Translating your argument into clearer language
Keep for yourself
  • What your argument actually is
  • Why you believe it
  • What evidence you find convincing and why
  • Where you're genuinely uncertain
  • What your original contribution is
Think together with Claude
  • Testing whether your argument holds under pressure
  • Exploring the strongest version of the counterargument
  • Deciding which of several directions to pursue
  • Figuring out what you actually think about something complex
KEY INSIGHT
Treat every Claude output as a first draft. It’s always the starting point for your thinking.

Thinking & Cognitive Offload Checklist — Step-by-Step Instructions

Click the bright cyan blue arrow at the end of each box to get your step-by-step instructions.

Work through each item. Check them off as you go.

0 / 2 COMPLETED
STEP 1 · The Interrogation — Push Back on Everything
~20 min
Stress-test Claude’s answer until your specifics appear
Force Claude past its general, smart-sounding first answer by confronting it with your actual thesis reality.
STEP 2 · The Ownership Statement — Make It Yours
~20 min
Write your position, then let Claude sharpen — not replace — it
Close Claude. Write five minutes by hand. Then use Claude only to sharpen what you already wrote.
KEY INSIGHT
Claude will always sound confident. Your job is not to match that confidence — it’s to know which parts of the confidence are earned and which parts are borrowed. The ones that are earned are yours. The ones that are borrowed need one more round.

Thinking Prompts for Thesis Students Working With Claude

A catalog of prompts you can reuse any time your thinking starts to drift or Claude sounds too polished. Copy any of them and adapt the brackets to your own thesis.

Category 1: Argument Stress-Testing

Purpose: Force yourself to defend your position before Claude softens or validates it.

What is the single strongest argument against my thesis? Don’t soften it. Make it as damaging as possible, then tell me if my current argument survives it.
Where is my argument most vulnerable? Identify the three points where a hostile examiner would attack first.
What am I assuming to be true that I haven’t actually proven yet?
If my thesis is wrong, what would that world look like? What evidence would I expect to find?
What is the weakest link in my reasoning chain? If that link breaks, what happens to the rest of my argument?

Category 2: Forcing Original Thinking

Purpose: Stop yourself from accepting Claude’s framing — generate first.

Before you answer, I want to share my thinking first: [your view]. Now tell me where my thinking is incomplete, not where it’s wrong.
I’m going to give you my rough, unpolished thinking on this. Your job is to help me develop MY argument further — not replace it with a better one. Here it is: [paste thinking].
What question should I be asking that I’m not asking yet?
What would a scholar who completely disagrees with my framework say about my methodology? Don’t resolve the tension — just surface it clearly.

Category 3: Exposing Hidden Assumptions

Purpose: Surface the thinking underneath the thinking.

What does my argument have to assume about [key concept] for it to hold? Are those assumptions stated or hidden?
What disciplinary assumptions am I bringing to this that a researcher from a different field would immediately question?
What am I treating as obvious that isn’t actually obvious?
If I changed just one foundational assumption in my framework, how would my entire argument shift?

Category 4: Productive Confusion

Purpose: Use Claude to name what you don’t yet understand, rather than paper over it.

I’m confused about [topic]. Don’t resolve my confusion yet — help me articulate exactly what I don’t understand. Make my confusion more precise.
I have two ideas that feel contradictory: [idea A] and [idea B]. Don’t reconcile them for me. Tell me what the actual tension is and why it matters.
What are the three most contested claims in my literature review? What is actually at stake in each debate?
Where in my argument am I using vague language to hide thinking I haven’t done yet?

Category 5: Deepening, Not Replacing

Purpose: Use Claude as a thinking partner that goes further, not sideways.

I want to go deeper on this idea, not broader. Stay inside this specific point and help me develop it further: [paste point].
What is the most intellectually interesting version of the argument I’m making? Push it further than I’ve taken it.
What are the second-order implications of my argument that I haven’t addressed yet?
If this argument is right, what else has to be true that hasn’t been examined yet?

What Else Can You Do with Prompts

  • Drafting Your Research Proposal
  • Pressure-Testing Your Thesis Argument
  • Writing Conference Abstract Drafts
  • Generating Advisor Meeting Agendas
● UP NEXT
Next: Session 02 — Your Thesis Operating System

In Session 02 you’ll set up 4 Claude Projects as a durable thesis workspace — with custom instructions, a curated knowledge base, and one Project per mode of thinking. The thinking, prompting & context curation skills you built today are what make every Project conversation about your thesis count.

← Session 00AUTO-SAVED · DRAFT