Lifelong Learning Club

Lifelong Learning Club

Share this post

Lifelong Learning Club
Lifelong Learning Club
How to Use Large Language Models to Design a Learning Path that Works

How to Use Large Language Models to Design a Learning Path that Works

If learning feels frustrating, you’re probably focusing on the wrong things. Here’s how to fix it.

Eva Keiffenheim MSc's avatar
Eva Keiffenheim MSc
Feb 24, 2025
∙ Paid
1

Share this post

Lifelong Learning Club
Lifelong Learning Club
How to Use Large Language Models to Design a Learning Path that Works
Share
Art by Anni Roenkae

“I quit.” The words surprised me. The person in front of me wasn’t the type to give up easily—she’d built a tech company from the ground up, grinding through all the challenges of startup life.

The rain still poured, trapping us inside the gym.

Like all good conversations on this island I’m currently working from, ours started with Burning Man, dancing, and community-building. Then DJing came up.

"I'm playing a set for a sober outdoor day dance tomorrow," I said.

"If the weather allows," we both added at the same time, laughing.

Her face lit up. “I actually gave DJing a real shot last year.”

Twelve hours of deep practice. Three lessons. Four hours each. Then he quit.

Not because she wasn’t talented. Not because she lacked motivation.

But because she focused on the wrong things.

“Why?” I asked.

She sighed. “I wanted to create a vibe. Read a crowd. Make people move. But instead, I was drowning in track organization, and software settings. It felt like I was training to be an accountant, not a DJ.”

I nodded and asked, “How many more hours do you think it would’ve taken to get past that?”

She thought for a second. “I don’t know… a lot?”

"Probably 30 more," I said. "Maybe less. But only if you focused on the right things."

Her eyes lit up. “How?”

Most people don’t quit because they’re incapable. They quit because learning feels overwhelming.

🔹 Your brain has limited working memory—too much at once, and progress stalls.

🔹 Cognitive Load Theory explains why some learning experiences feel effortless while others feel impossible.

🔹 Without a system to filter essential skills from distractions, frustration builds—and motivation dies.

This isn’t just about DJing. It’s why people abandon fitness goals, struggle to write better or give up on learning a new language.

The solution? Reduce cognitive overload by structuring your learning around high-leverage sub-skills—the 20% that drive 80% of results.

In this article, you’ll discover a simple framework to break down any skill into essential, high-impact sub-skills and how to leverage LLMs like Grok3 to design a 20-hour learning plan that prioritizes real progress.

Most people think learning is about putting in more time. It’s not. It’s about putting time into the right things.


Why People Get Stuck (And How to Fix It)

Most people think learning is about putting in more time. It’s not.

The real challenge isn’t effort—it’s cognitive overload.

When you’re learning something new, your brain is juggling different types of information. If you don’t manage cognitive load, even the most exciting skill can feel exhausting. That’s when frustration creeps in, motivation tanks, and you start telling yourself:

"Maybe I’m just not good at this."

But quitting has nothing to do with talent. It has everything to do with how you structure your learning.

🔹 Your attention doesn’t disappear—it just gets spread too thin.

🔹 Cognitive Load Theory explains why some learning experiences feel effortless while others feel overwhelming.

🔹 The fix? Reduce cognitive load by structuring your learning around high-leverage subskills—the critical few that unlock everything else.


Cognitive Load Theory in Action

Most people unknowingly overload their cognitive capacity when learning something new. Instead of strategically filtering what’s essential, they try to grasp everything at once. The result? Frustration, overwhelm, and quitting too soon.

There are three types of cognitive load at play:

  1. Intrinsic Cognitive Load – The actual difficulty of a task.


    👉 Hack: You can’t change how complex a subject is, but you can control when you encounter difficulty. Start with foundational concepts before diving into advanced material. Example: If you’re learning jazz piano, don’t begin with improvisation—first master chord progressions.

  2. Extraneous Cognitive Load – Unnecessary friction in the learning process.


    👉 Hack: Simplify your environment. Reduce distractions. Use well-designed instructional materials. Example: Learning from a poorly structured book? Switch to a course that breaks things down visually.

  3. Germane Cognitive Load – The effort needed to store new knowledge into long-term memory.


    👉 Hack: Stop passively consuming. Convert learning into action. Example: Instead of rewatching tutorials, force yourself to use the knowledge—summarize key insights, apply them immediately, crate Anki cards, or teach someone else.

AI as a Learning GPS: Stop Wasting Time on the Wrong Things

Most people don’t struggle with motivation. They struggle with not knowing what to focus on.

This is where AI changes the game. Instead of blindly guessing what to study next, let AI map out the most effective learning path for you.

Step 1: Define Your Goal with Precision

AI is only as useful as the clarity of your input. Try refining your learning goals:

❌ Weak Goal: “I want to run more.”
🔹 Why it's weak: Too vague—how far, how often, and to what end? There’s no way to track progress or measure success.
✅ What do instead: “I want to run a half-marathon in four months, finishing in under two hours—while avoiding injuries and genuinely enjoying the training process.”

❌ Weak Goal: “I want to tell better stories.”
🔹 Why it's weak: Improvement is subjective—what kind of storytelling? In what format? Without a concrete application (emails, presentations, etc.), it’s directionless.
✅ What do instead: “I want to learn and apply storytelling techniques to structure my emails, presentations, and pitches in a way that increases engagement and buy-in from my team.”

❌ Weak Goal: “I want to be a better teacher.”
🔹 Why it's weak: Teaching is broad—does this mean better student engagement, clearer explanations, or improved retention rates? Without specificity, it’s impossible to focus on actionable improvement.
✅ What do instead: “I want to learn how to design interactive lectures using cognitive science principles—so my students retain at least 50% more of the material and actively engage in class discussions without relying on passive note-taking.”

A clear goal leads to better learning recommendations.


Step 2: Ask AI to Identify Key Subskills

Now, let AI do the heavy lifting. Try this prompt:

📝 AI Prompt 1

Keep reading with a 7-day free trial

Subscribe to Lifelong Learning Club to keep reading this post and get 7 days of free access to the full post archives.

Already a paid subscriber? Sign in
© 2025 Eva Keiffenheim
Privacy ∙ Terms ∙ Collection notice
Start writingGet the app
Substack is the home for great culture

Share