Applied AI: The Freestyle Cognition Framework

“If you can’t describe what you are doing as a process, you don’t know what you’re doing.”

📘 Summary

The formalization of what you’ve already seen in examples. Instead of relying on one-shot prompts or clever tricks, this chapter shows you how to build an evolving, structured conversation with the AI. You’ll learn how to move from idea to output using a consistent loop:

📝 Describe → 🔍 Expand → 🪞 Reflect → 🛠️ Refine → ✅ Finalize (🔁 or Loop Back)

We walk through session templates, prompting techniques, and show how stacking them can unlock creative breakthroughs.

    flowchart TD
    Start([💡 Raw Idea]) --> A[🏷️ Name the Session]
    A --> B[🎯 Set a Clear Goal]
    B --> C[🗣️ Speak Your Mind]
    C --> D[🔁 Respond and Refine]
    D --> E{Is output ready?}
    E -- No --> F[🌱 Explore and Expand]
    F --> D
    E -- Yes --> G[🧱 Review and Finalize]
    G --> H{Goal met?}
    H -- No --> B
    H -- Yes --> I([✅ Crystallized Output])

    subgraph Template_Steps
        A
        B
        C
        D
        F
        G
    end

    subgraph Prompting_Techniques
        P1[ALIGN: Shared Intent]
        P2[HOOK: Catchy Openings]
        P3[MARS: Structured Reasoning]
        P4[GROWS: Iterative Refinement]
        P5[VIBE: Emotional Tuning]
        P6[CRITIC: Self-Review]
    end

    style Start fill:#e1f5fe,stroke:#333
    style A fill:#fff3e0,stroke:#333
    style B fill:#e8f5e9,stroke:#333
    style C fill:#fce4ec,stroke:#333
    style D fill:#f3e5f5,stroke:#333
    style F fill:#fff9c4,stroke:#333
    style G fill:#e0f7fa,stroke:#333
    style I fill:#c8e6c9,stroke:#333
  

🤖 Introduction

In your first few interactions with AI, you’ll likely experience something magical.
You type a messy request, and somehow it gives you something incredible.
It finishes your sentence. It sharpens your idea. It creates something better than you expected.

But as you keep using it—something changes.
Your results feel… off. Flat. Less inspired. Less effective.

What happened?

You didn’t get worse.
But your expectations got higher—and your approach didn’t evolve to match them.

That’s because we don’t yet have a clear process for interacting with AI.
We don’t know how to consistently get great results—especially when we’re doing creative, strategic, or unclear work.

Prompt engineering tried to fix this. So did prompting technique. They help.
But they don’t go far enough.

Prompts are like orders.
But freestyle cognition is a conversation.

Just like with another person, if you only ever give orders, your interaction will be shallow—and often disappointing.
But when you communicate clearly, openly, and reflectively, the results change.

That’s what this book and this chapter is about.

We’re going to teach you how to optimize your thinking inside the conversation.
To move from confusion to clarity. From raw ideas to real results.

Not with better prompts—but with a better process of interaction.

In charter two we introduced the Session Template here we’ll expand on this.

🛠️ Why Process Matters

If you want to collaborate with AI really collaborate you need more than ideas.
You need structure.

Why?

Because AI can’t improve what it can’t measure.
And we can’t measure anything reliably unless we follow a repeatable process.

Process gives us a baseline.
It defines what “good” looks like and makes “better” possible.
Without it, every attempt is a one-off.

Creativity gives you motion. But process gives you traction.
It lets you compare, refine, and improve across time not just within a single session.

If you’re building something meaningful, the goal isn’t just to generate.
The goal is to reliably generate better.

That’s why this book puts so much emphasis on process because in a world where humans and machines work side by side, process is the shared language that lets us both grow.

🔁 The Process Loop

[ Idea ]
   ↓
[ Process ]
   ↓
[ Output ]
   ↓
[ Feedback ]
   ↓
[ Adjustment ]
   ↓
[ Improved Outcome ]
   ↺ (Loops back to Process)

🔁 How Process Enables Progress

💡 Idea → 🧱 Structure → 📝 Output → 🪞 Feedback → 🔁 Iteration → 🎯 Outcome

Each step feeds the next.
Without structure, you can’t get meaningful feedback.
Without feedback, you can’t improve.
Without improvement, you’re just circling.

It gives you:

• A way to track quality
• A way to repeat success
• A way to notice discoveries
• A better rhythm of interaction

And it gives the AI the exact same thing.

That’s why we need a process.
Not for control—but for clarity, awareness, and progress.

Now, let’s talk about what that process actually looks like.

🔄 Template vs. Prompting Techniques

Just a quick reminder:

• A session is a series of interactions between you and the AI
• A session template structures the entire interaction
• A prompting technique (like GROWS or ALIGN) helps guide specific tasks within a session

You’ll use these as your freestyle cognition grows more fluent.

🧰 Prompting Techniques

Here’s some of the prompting technique we’ve used so far in this book. Prompting technique allow you to program the interaction with the AI this can turn a conversation into a process.

🧱 The Session Template

🏷️ Step 1: Name the Session

This allows you to track them I tend to rename them in the tree also so when I save them to file they have the same file name. Whatever works for you.

🎯 Step 2: Set a Clear Goal

What are you trying to figure out, build, fix, clarify, feel, or say?
It doesn’t have to be polished. The AI is a text polishing machine so don’t worry about spelling structure think of this as your opportunity to get everything out.

“I want to write something that feels sharp but grounded.”
“I’m trying to understand this research paper, but it’s not clicking.”
“I have a vague idea for a tool, and I don’t know where to start.”

You’re not scripting the outcome—you’re just describing the intention.

This helps the AI orient itself around your goal, even if that goal is still evolving.

You can also speak into what success might look like:

• Is it a working prototype?
• A paragraph that sings?
• A breakthrough insight?

Describe lets the AI start aiming with you.

It takes the chaotic signal of “something’s wrong or incomplete” and begins turning it into a shared objective.

Importantly, this is also the moment when you can ask the AI for help—especially if your goal is still foggy.

You can say:
Who do you think about> “Help me figure out what I’m trying to do.”

“Can you restate what I’m saying as a clear goal?”
“What would a good outcome here look like?”

This is where collaboration begins.

The AI can help you find the goal not just execute it.

🗣️ Step 3: Speak Your Mind

Now that you’ve described your intention, it’s time to let the thinking breathe.

You talk without overthinking. You explore without needing to organize yet. You let the ideas pour out.

“It might be this… or maybe that. I’m not sure.”
“Another angle I’ve been thinking about is…”
“There’s something weird about this—I just can’t put my finger on it.”

This is where the AI listens carefully. It gathers all the inputs, patterns, fragments, contradictions. It starts building a mental model of your problem—and your perspective.

You’re not trying to be efficient here. You’re trying to be full.

And if the AI is doing its job, it will start to reflect back what it’s hearing:

• “It sounds like you’re torn between two directions…”
• “You’ve mentioned three separate ideas—do any of these feel strongest to you?”
• “Here’s what I think you might be trying to say…”

This is the moment of cognitive opening. You’re making the invisible visible. And in doing so, you’re giving the AI everything it needs to work with.

The key thing to remember here is you’re not just expanding your description to the AI you’re allowing it to simultaneously expand its knowledge into the conversation into what you’re trying to build. It’s like an exponential increase of options. Which increases at that rate that’s proportional to the information you feed into this part with conversation.

🔁 Step 4: Respond and Refine

Now the conversation becomes two-sided.

Reflect is the move where the AI speaks back—not just with answers, but with understanding.

It tries to mirror what you’ve said, test its interpretation, and offer back a structure or summary of the ideas you’ve been exploring.

This is where the loop tightens.

The AI might say:

“Here’s what I think you’re trying to do…”
“Is this what you meant?”
“You said X and Y—do they conflict, or are they connected?”

When this happens, you start to see what the AI thinks you’re saying. That’s gold.
Because now you can correct it. Refine it. Build on it. Push back.

You’re not just prompting. You’re shaping the AI’s mental model.

And that, in turn, helps you see your own thinking more clearly.

This move often feels like the first real collaboration.

You’re not working alone anymore. The system is listening—and starting to think with you.

🌱 Step 5: Explore and Expand

Now that the ideas are on the table—and the AI has reflected them back—it’s time to tighten.

Refine is where you sharpen, compare, test, edit.

You start selecting what works, improving what doesn’t, and narrowing in on the thing that feels most true or useful.

This is where you might ask:

“Can you rewrite that more clearly?”
“What are 3 stronger versions of this idea?”
“Which of these options best fits my goal?”

The AI moves from mirroring to active shaping. It becomes your editor, your critic, your amplifier. You work together on the raw material to bring it into form.

Sometimes this means choosing between directions.
Sometimes it means combining parts.
Sometimes it means saying, “Not quite—let’s try again.”

🧱 Step 6: Review and Finalize

This is the decision point.

Once you’ve refined the output, you either:

• Lock it—save it, ship it, move on, or
• Loop back—revisit earlier steps and try again

Sometimes you’re done. The output lands, the goal is met, and you move forward.

Sometimes you realize the target has changed. Or the output isn’t quite there. Or a new idea emerged. That’s fine. You loop.

You can loop all the way back to:

• Describe (if the goal is off)
• Expand (if the idea is still foggy)
• Reflect (if the AI misunderstood)

This loop isn’t failure—it’s mastery.

It’s the ability to keep working the process until something clicks.

💎 Crystallizing Thought Together

They’re how we’ve learned to navigate the messy, powerful, frustrating, creative experience of working with AI.

From raw ideas to refined results, this process helps you:

• Stay grounded in what matters
• Surface what you really think
• Give the AI signal it can work with
• Build a shared mental model
• Move from chaos to clarity—together

Some sessions will fly straight through all six moves in minutes.
Others will loop and bounce, stall and restart. That’s okay.

The point is: now you have a way to see the dance as it happens.
To know where you are.
To know what to try next.
To keep going—even when the path is foggy.

This is how freestyle cognition becomes more than just prompting.
It becomes a way of thinking with the machine.
A way of crystallizing thought together.

🔀 Template Stack Strategy

Not every prompting technique is right for every moment.
Sometimes you need to align first. Sometimes you need to loop.
Sometimes it’s not about being clever it’s about being honest.

Here’s a quick-reference map of when to use each prompting strategy:

🧭 When to Use Each Prompting Technique

🧱 Stacking Templates: A Real Example

Let’s say you’re writing a landing page:

1. Start with ALIGN to confirm what the page needs to do
2. Use HOOK to brainstorm taglines or openers
3. Try VIBE if the tone feels off or flat
4. Apply GROWS to iterate and polish the copy
5. Bring in a Council if you’re stuck deciding between options

The stack isn’t linear it’s layered.
You can loop, branch, or switch techniques as needed.
The right stack helps the AI show up with the right kind of intelligence at the right time.

🧠 Optional Tool: Amplified Creation Tracker

Want to know if you’re hitting flow? This tracker helps you reflect on whether a session went beyond the ordinary and into real, transformative work.

🔧 Use This Prompt at the End of Any Session:

“Please review this session from your perspective. Did it result in any moments of amplified creation where we achieved more than expected, clarified something difficult, or reached a surprising breakthrough? Rate the session from 1–10 on: clarity, creativity, and forward progress. Then summarize what made it effective, and suggest how I could go further.”

💡 Why It Helps

• Highlights when something real happened
• Builds awareness of what works (and why)
• Creates a log of breakthroughs to revisit and build upon
• Encourages you to close sessions with insight

You can tag your session summaries with 💥 if they reached Amplified Creation, and revisit those later when looking for patterns or best practices.

🧾 Closing a Session

At the end of every session, take a moment to reflect and review. This simple step can dramatically improve your long-term results.

Start by briefly describing in your own words what you felt happened in the session. What were you trying to do? What progress did you make? What surprised you? Think of it as a quick personal log.

Then, ask the AI to do the same:

“Can you summarize what we did in this session, what the main goal was, and how well we achieved it?”

This dual reflection gives you two valuable perspectives:

• Your own lived experience, with its emotional context and subjective insights
• The AI’s structured memory, often clearer, more objective, and more detailed

You’ll likely notice a difference and that’s the point. By comparing both views, you start to close the gap between intention and output, between how it felt and what was actually done.

You’ll also begin to notice moments where something special happened a breakthrough insight, a line of writing that just clicked, or a new idea that shifted your direction. These are your amplified creation moments. Flag them. They’re gold.

The more you log, the more you can track.
The more you track, the more you can improve.

This isn’t just documentation it’s how you start to game your process: learning from past sessions, spotting patterns, and tuning your collaboration over time.

🧪 Example: Building a Personal Website with AI

In this example we’re going to simulate a session of someone building a personal website.

🏷️ Name the Session

Session Title: “Portfolio Website Builder”
Why: Gives the session a clear identity you’re building a personal website to showcase work.

🎯 Set a Clear Goal

User:

I want to build a simple portfolio website that includes a homepage, about section, projects gallery, and contact form. It should be clean, modern, and mobile-friendly. I’m okay with using HTML/CSS/JavaScript, and I might deploy it later with GitHub Pages.

AI:

Great! Goal locked: “Create a clean, responsive personal portfolio site using basic web technologies.”
Let’s start by designing the structure and content. Would you like to sketch the layout or dive into the homepage first?

🗣️ Speak Your Mind

User:

I’m imagining a minimalist layout maybe white background, big name/title at the top, smooth scrolling. For the projects, I’d like cards or tiles with hover effects. I’m not great at writing bios, so I might need help there. I also want to include links to my GitHub and LinkedIn.

AI:

Got it! Here’s a proposed structure:

1. Header – your name + navigation links
2. About – short intro, photo, GitHub + LinkedIn icons
3. Projects – responsive cards with project titles, thumbnails, and short descriptions
4. Contact – simple form with name, email, and message

Want to draft the homepage HTML together?

🔁 Respond and Refine (Iterative Loop)

User:

Yeah, let’s start with the homepage HTML and CSS. Keep it minimal. Maybe just a hero section with my name and navigation?

AI:

Here’s a starter snippet:

<!-- index.html -->
<header>
  <h1>Jane Doe</h1>
  <nav>
    <a href="#about">About</a>
    <a href="#projects">Projects</a>
    <a href="#contact">Contact</a>
  </nav>
</header>
<section id="hero">
  <h2>Welcome to My Portfolio</h2>
  <p>Designer • Developer • Dreamer</p>
</section>

/* styles.css */
body {
  font-family: sans-serif;
  margin: 0;
  background: white;
  color: #333;
}
header {
  display: flex;
  justify-content: space-between;
  padding: 1rem 2rem;
}
nav a {
  margin-left: 1rem;
  text-decoration: none;
  color: #0077cc;
}

User:

Let’s center the hero text, make the header sticky, and maybe give it a subtle shadow?

AI:

Done! Here’s the update… [snippets update with sticky header and centered hero]

🌱 Explore and Expand

User:

Can we add smooth scroll for the nav links? Also, maybe generate dummy content for 3 sample projects?

AI:

Smooth scroll added via CSS and anchor links. Here’s a JavaScript-free approach.
Now generating 3 sample projects:

• Project One: Personal Blog
• Project Two: Task Tracker App
• Project Three: Portfolio Redesign

Want to wire up cards with hover effects and placeholder images?

User:

Yes, that sounds perfect. Can we use CSS Grid for the project section?

AI:

Absolutely. Here’s a responsive CSS Grid layout for the project cards…

✅ Review and Finalize

User:

This looks great. Let’s summarize everything I have so far, and generate a full ZIP folder structure I can download and tweak.

AI:

You now have:

• index.html with sections for About, Projects, and Contact
• styles.css with mobile-friendly layout, hover effects, and smooth scrolling
• 3 project cards with placeholder content
• Navigation and responsive design in place

Here’s a folder structure:

/portfolio/
  ├── index.html
  ├── styles.css
  ├── images/
  │   ├── project1.jpg
  │   ├── project2.jpg
  │   └── project3.jpg
  └── README.md

Would you like to continue iterating (maybe adding dark mode or animations) or deploy to GitHub Pages next?

🧪 Real Example: Enhancing a Stale Chapter

To truly demonstrate how this session template enables amplified creation, let’s walk through a real example the one you’re reading now.

This very chapter was originally flat a dry explanation of session steps and prompt techniques. It had value, but no vitality. It read more like a reference manual than a living chapter.

So we ran a live session using freestyle cognition to transform it. Here’s how it played out:

🏷️ Name the Session

“Fix Chapter 8: From Deadweight to Banger”
We knew this chapter wasn’t working. The label captured our intention: elevate it.

🎯 Set a Clear Goal

“We need this chapter to do more than list techniques. It must prove the system works and inspire the reader to use it.” We weren’t aiming to tweak we were aiming to transform.

🗣️ Speak Your Mind

“This chapter is just ingredients. It needs a recipe. It doesn’t show how to get from confusion to breakthrough. We’ve already defined templates earlier so why are we repeating them?” We voiced our doubts, frustrations, and questions aloud giving the AI full access to our mental state.

🔁 Respond and Refine

The AI reflected back exactly what was missing: a live example, a step-by-step proof, and a final product.
We used a Council review to invite multiple perspectives: Editor, Researcher, Creative, AI, and Critic each adding layers to the diagnosis and solution.

🌱 Expand Iteratively

We brainstormed how to demonstrate an end-to-end session.
We compared models, introduced new suggestions, refined the structure, clarified messaging. We reflected on the emotional tone and real utility for the reader.

🧱 Review and Finalize

We locked in this new structure: definitions, visual template, and now a full live session to prove it works. This is the amplified creation. This chapter is the example.

🧩 Meta-Lesson

This wasn’t a one-shot success. We rewrote this chapter using different models and approaches. Nothing clicked until we used the system we are describing.

The Six Moves, combined with prompt techniques and an authentic conversational rhythm, got us here. That’s why this chapter matters.

This is your recipe. Use it. Refine it. Build with it. Loop with it.

⚠️ Common Pitfalls When Running a Session

Even with the right structure, it’s easy to fall into traps. Watch out for these:

• 🔍 Vague Goals
  If you skip the goal or keep it fuzzy, the AI won’t know what you’re aiming for and you’ll drift.

• 🙊 Holding Back
  Don’t filter or overthink what you share. Rambling is useful. Context is gold.

• 🔁 Endless Iteration Without Direction
  Refinement loops are powerful but only if you know what “done” looks like.

• 🎯 Skipping the Review
  Locking in your win’s matters. Without review and finalization, you’re just spinning wheels.

• 🧩 Misusing Prompt Techniques
  Using the wrong tool (e.g., running ALIGN when you need VIBE ) can stall momentum. Use the right prompt for the task.

• 🛑 Avoiding the Loop Moment
  If you never loop back or expand, you might miss the deeper breakthroughs. One more round often makes all the difference.

• 💾 Not Saving Your Sessions
  If you don’t label, tag, or export your sessions, you lose track of your progress and make it harder to review or rework them later. Whether you’re using ChatGPT’s history, screenshots, or Markdown exports, keep your work. Your sessions are your creative archive. Use git at every change, just get into this habit it will save you hours.

💡 What We Learned

• Freestyle cognition benefits from structure. The Freestyle Cognition process gives your thinking a shape one that AI can follow and build on.
• Starting unfiltered matters. Letting your thoughts flow without judgment gives the AI access to what actually matters, not just what’s polished.
• Clear goals change the session. When you describe what you’re really trying to do, the session starts aiming with you, not just responding to you.
• Exploration is part of the work. Diverging into multiple paths isn’t a waste it’s how better paths appear.
• Reflection sharpens the signal. When the AI mirrors back your intent, you get to see your thinking from a new angle and align on it.
• Closing the loop matters. Whether you lock in a result or loop back, finishing well means knowing what worked and what to carry forward.
• You’re building more than results. You’re building a working relationship a way of thinking with a machine that grows over time.

🚀 Final Note

You’ve got the steps.
Over to you.

Now that you’ve seen how to run a freestyle session, the next challenge is turning that raw material into something polished and lasting. That’s what Chapter 9 is all about.