Turn a YouTube course into a mini learning app using NotebookLM + Gemini. Zero code.

Stop watching tutorials passively. Turn them into tutors, coaches, and practice apps that teach you back.

Most people use YouTube to learn. But the problem is simple:

You watch a course. You feel productive. Then you forget most of it.

I found a better workflow:

YouTube course → NotebookLM → Gemini Canvas → mini learning app

Instead of just watching a video, you can turn that video into an interactive tutor.

For example:

• A SQL course becomes a SQL coach
• An interview video becomes an interview prep app
• A language course becomes a speaking tutor
• A history lecture becomes a personal history teacher
• A book talk becomes a book tutor

No coding needed. Just NotebookLM, Gemini, and one good prompt.

The Problem With Learning From YouTube

YouTube is one of the best free learning platforms.

You can find courses on almost anything:

• SQL
• DSA
• AI engineering
• System design
• History
• Languages
• Business
• Writing
• Coding

But most people learn passively. They watch videos.

Maybe they take notes. Maybe they save the playlist.

But they don’t actively practice.

That is where learning breaks.

Because real learning does not happen when you only consume.

It happens when you:

• Answer questions
• Make mistakes
• Get feedback
• Revise weak areas
• Apply the concept
• Repeat the process

This is why I like this workflow.

It turns a passive video into an active learning system.

The Simple Workflow

The idea is very simple.

Use NotebookLM as the knowledge base. Use Gemini Canvas as the app builder.

NotebookLM understands the YouTube video. Gemini turns that understanding into an interactive app.

Think of it like this:

NotebookLM = brain & Gemini Canvas = interface/app

One stores the course knowledge. The other builds the learning experience around it.

By the end, you can build apps with NotebookLLM even from a YouTube video.

In this post, I built a Spanish tutor, a history professor, an interview coach, and a few more. Some of them replace a good old subscription that costs hundreds a year. Once you realize this, we say R.I.P to a lot of popular apps.

Note: If you are new to this, two questions are sitting in your head right now. How do I train NotebookLM? And how do I turn it into an app?

Let us answer those first.

How NotebookLM and Gemini Work Together

Two tools, one job each NotebookLM holds the knowledge Gemini builds the app. One feeds the other.

We’ll do it in 3 steps.

Step 1: Add a YouTube Course to NotebookLM

First, choose a YouTube video or playlist you want to learn from.

It can be:

• A long course
• A university lecture
• A tutorial
• A playlist

Play

Then:

1. Open NotebookLM
2. Create a new notebook
3. Add a source
4. Paste the YouTube link
5. Let NotebookLM read the transcript and understand the content

Now NotebookLM can reason over that video.

Instead of you manually taking every note, NotebookLM can help summarize, explain, and structure the material.

Step 2: Connect NotebookLM With Gemini

Now open Gemini.

Then:

Play

1. Click the plus button
2. Add your NotebookLM notebook as a source
3. Open Canvas
4. Ask Gemini to build an app from the notebook

This is the step many people miss.

They use NotebookLM only for summaries.

But the more interesting use case is:

turning source material into an interactive learning tool.

Step 3: Write One Clear Prompt

You don’t need a complex prompt.

You just need to describe the app you want.

Play

Example:

Build a tutor from this notebook that teaches me one concept at a time, asks me questions, checks my answers, explains mistakes, and gives me practice based on my weak areas.

That’s it.

Gemini can use the NotebookLM source and create an app-like experience inside Canvas.

Now the course is no longer just a video.

It becomes a tutor.

5 App Ideas You Can Build From YouTube Courses

Here are 5 simple examples I’m exploring.

1. SQL Coach

If you are preparing for placements or interviews, SQL is a perfect use case.

Take a good SQL YouTube course and add it to NotebookLM.

Then use this prompt:

Build a SQL coach from this notebook.

It should teach one concept at a time, generate practice questions, review my SQL answers, explain my mistakes, and give me adaptive exercises based on my weak areas.

Now instead of only watching SQL tutorials, you can practice.

The app can:

• Explain SELECT, JOIN, GROUP BY, HAVING, subqueries, etc.
• Ask interview-style questions
• Check your answers
• Give hints
• Suggest what to revise next

This is much better than passive learning.

2. Interview Prep Coach

This one is very useful for students.

Take a YouTube video on AI engineer interviews, software engineering interviews, system design, or HR questions.

Add it to NotebookLM.

Then use this prompt:

Build an interview prep coach from this notebook.

It should ask me one interview question at a time, let me answer, score my response out of 10, explain what was missing, and show me how to improve.

You can also add:

Create two modes:
1. Text mode
2. Voice practice mode

In voice mode, let me speak my answer and receive feedback.

This can help with:

• Technical interviews
• HR interviews
• AI/ML interviews
• SQL interviews
• System design discussions

For someone preparing for placements, this is a very practical workflow.

3. DSA Tutor

You can also use this for DSA.

Take a YouTube video on arrays, recursion, dynamic programming, graphs, trees, or sliding window.

Then prompt:

Build a DSA tutor from this notebook.

It should explain one concept at a time, ask coding interview questions, give hints before showing the solution, evaluate my approach, and explain time and space complexity.

This is useful because DSA is not just about watching solutions.

You need to practice thinking.

A DSA tutor can help you:

• Understand patterns
• Get hints
• Compare approaches
• Learn complexity analysis
• Practice weak topics

4. Language Tutor

You can turn a language course into a tutor.

For example, if you want to learn Spanish, French, German, or Japanese, add a beginner YouTube course to NotebookLM.

Prompt:

Build a language tutor from this notebook.

It should give me one short lesson at a time, ask me to translate simple sentences, correct my mistakes in plain English, and track what I need to practice again.

You can also add:

Add a speaking practice section where I can speak and receive feedback using browser voice features.

This turns a normal language video into an interactive learning experience.

5. Book Tutor

This is also interesting.

If an author has a TED Talk, lecture, interview, or summary video about a book, you can use that as a learning source.

For example, if you want to understand a business, psychology, or marketing book, add a related talk to NotebookLM.

Prompt:

Build a book tutor from this notebook.

It should explain the core ideas, ask me how I would apply them, challenge my answer, and give examples from real life.

This is useful because good books are not just meant to be summarized.

You need to apply the ideas.

A book tutor can help you think deeper.

Why This Workflow Matters

The important part is not just that you can “build apps.”

The important part is this:

AI can turn learning content into active practice.

That is a big shift.

Earlier:

Watch video → take notes → maybe forget

Now:

Watch video → convert to source → build tutor → practice actively

This is much more powerful.

Because the app can:

• Ask you questions
• Check your understanding
• Give feedback
• Adapt to your weak areas
• Make you apply the concept
• Help you revise faster

That is how learning should work.

The Pattern

The pattern is simple:

1. Pick a YouTube course
2. Add it to NotebookLM
3. Connect it to Gemini
4. Open Canvas
5. Ask Gemini to build a tutor/coach/app

One video becomes one learning app.

One playlist becomes one complete coach.

One topic becomes one practice system.

My Favorite Use Case

For me, the most useful version is:

Placement prep coach

Because I can use it for:

• SQL
• DSA
• AI interview prep
• System design basics
• HR questions
• Resume-based mock interviews

Instead of randomly watching more videos, I can turn the same videos into practice tools.

That is a better workflow.

A Few Things To Remember

This is not perfect.

You still need to verify information.

You still need to practice seriously.

You still need to use good sources.

A weak YouTube video will create a weak learning app.

So choose sources carefully.

Better sources:

• University lectures
• Official tutorials
• Well-known educators
• High-quality long-form courses
• Documentation-based videos
• Verified technical content

Also, don’t blindly trust every answer.

Use AI as a learning assistant, not as the final authority.

Prompt Template You Can Reuse

Here is a simple reusable prompt:

Build an interactive learning app from this notebook.

The app should:
1. Teach one concept at a time
2. Ask me questions after each concept
3. Check my answers
4. Explain my mistakes clearly
5. Give me practice exercises
6. Track my weak areas
7. Suggest what I should revise next

Keep the interface simple and beginner-friendly.

For interview prep, use this:

Build an interview coach from this notebook.

It should ask one question at a time, wait for my answer, score it out of 10, explain what was good, explain what was missing, and show me a better answer.

For coding/DSA:

Build a coding tutor from this notebook.

It should explain the concept, give coding problems, provide hints, review my approach, and explain time and space complexity.

Final Thought

Most people use AI to summarize content.

But the better use case is:

turn content into systems.

A YouTube video can become a tutor. A course can become a coach. A playlist can become a practice app. A lecture can become an interactive learning system.

That is the real 1% improvement.

Not just consuming more.

Learning better.

What I’ll Try Next

I’m going to test this workflow for my own placement prep.

My plan:

1. Take one SQL course
2. Turn it into a SQL coach
3. Take one DSA video
4. Turn it into a DSA tutor
5. Take one AI interview video
6. Turn it into an interview prep coach

Then I’ll share what actually worked and what failed.

Because this is the kind of AI workflow that can genuinely save time for students, developers, and builders.

Inspired by a NotebookLM + Gemini workflow I saw from Gencay

It’s obvious, but I’ll say it again.

I don’t care about Claude,Gemini or any other AI model. I don’t pick sides.

I’m sharing, ONCE a week, how my work is changing (very fast) with AI.

As I’m trying to keep up, I want you to keep up. so you stay ahead. Well, I will continue to do so for any future upgrades.

Because I want to be the greatest filter to the AI noise. And 2000+ people trust me to be their filter. ALL because of my LinkedIn. But most readers subscribed because someone they trusted sent them one of my articles.

If this helped you, share it with someone who’s still doing things the hard way. It takes less than a second, but it might genuinely change how they work.

And if someone forwarded this to you — thank them, you’re now part of the loop.

You can subscribe for free here to get the next one.