Stop Watching Tutorials. Do This Instead

Thanks to HubSpot for sponsoring this video. AI is developing at a rapid pace,

video. AI is developing at a rapid pace, and with it, new tools and technologies are dropping every [music] single week.

With the imminent loom of layoffs in the tech job market, the people who can update themselves quickly with new technologies are the ones that can control their career trajectory instead of reacting retroactively. [music]

However, lots of developers waste months learning new tech stacks the wrong way.

They binge tutorials, take endless courses, and still freeze when it's time to actually build something. Hi friends,

[music] I'm Maddie. I'm a senior software who previously worked at Google and interned at other big tech companies like Amazon, IBM, and Microsoft. Over

the years, while earning my bachelors at MIT, completing six internships, and two industry roles, I've had to rapidly learn new languages, frameworks, and tools, sometimes only weeks before the

project deadline. Today, I'm going to

project deadline. Today, I'm going to share the exact strategies that help me go from zero to productive in a new tech stack fast. [music]

stack fast. [music] We'll cover why most people stay stuck in tutorial hell, the sciencebacked learning techniques that actually work, and how to use AI to accelerate your learning like never before. Let's get

into it. We'll start with what might be holding you back. Tutorial hell.

Tutorial hell is when you watch course after course, follow along every video, feel like you're learning, but the moment you try to actually build something on your own, your mind goes blank. You've probably experienced this.

blank. You've probably experienced this.

You finish a React tutorial feeling confident and then you open VS Code to start your own project and just stare at the screen. The reason why this happens

the screen. The reason why this happens is because tutorials create an illusion of competence. When you're following

of competence. When you're following along, your brain isn't doing the heavy lifting. The instructor is. You're

lifting. The instructor is. You're

recognizing the patterns instead of actually knowing and internalizing it yourself. And recognition is way easier

yourself. And recognition is way easier than recall. Research in cognitive

than recall. Research in cognitive psychology calls this the fluency illusion. Since the material feels

illusion. Since the material feels familiar, you assume that you've learned it. But when there's no one guiding you

it. But when there's no one guiding you step by step, that familiarity vanishes.

I've definitely been so guilty of this before. For example, when doing le code

before. For example, when doing le code practice, I'll watch Le code walkthroughs and read the solutions of problems, assume that I know how to solve that problem. But when the same pattern shows up in a different problem,

I draw a blank. So, the fix for this is that instead of passively consuming, you need to actively struggle. And I'll show you exactly how to do that in the next few sections. If you want to go even

few sections. If you want to go even deeper on this, I want to share something that ties directly into what we're covering today. HubSpot put

together a free guide called Learn to Code with ChachiBT in collaboration with my friend Sundus Khaled and it's perfect for anyone trying to pick up new programming skills quickly. It shows you exactly how to use ChachiBT as your

personal programming mentor and walks you through how to leverage AI to answer your coding questions, debug your code when you're stuck, and accelerate your learning in ways that weren't possible even a few years ago. Inside, you'll

learn specific prompts and techniques to get useful explanations instead of generic responses. It covers how to

generic responses. It covers how to break down complex programming concepts, how to get AI to walk you through code line by line, and how to use it for real-time feedback while you're building projects. My favorite section is about

projects. My favorite section is about the common mistakes people make when learning with AI, and how to avoid them so you're actually retaining what you learn, not just copy and pasting.

Overall, the guide is super practical.

Whether you're a complete beginner or actively leveling up. It gives you the actual techniques to use AI effectively.

Check the link in the description to download your free copy and start coding smarter. Thanks to HubSpot for creating

smarter. Thanks to HubSpot for creating this resource and sponsoring this video.

All right, let's get back to the next tip. The second tip I have for you is to

tip. The second tip I have for you is to start building before you feel ready.

Most people think that they need to finish a comprehensive course before they can build anything meaningful. But

the best way to learn is to have a project that forces you to learn what you need when you need it. For example,

I was ramping up on a new framework at Google. I didn't spend weeks watching

Google. I didn't spend weeks watching tutorials. I read the getting started

tutorials. I read the getting started documentation and then found a small well-defined task, something that I could realistically finish in a few days, and I just started coding. Every

time I hit a wall, I look up that specific concept, implement it, and move on. This approach works because of

on. This approach works because of something called just in time learning.

Instead of trying to frontload everything into your brain and hoping that it sticks, you learn in context.

The information has immediate relevance so your brain actually retains it. So

here's what I recommend. Within the

first day or two of picking up a new stack, start a mini project. It doesn't

have to be impressive. Build a simple API, create a basic to-do app, clone a single feature from an app you use. The

point is to get your hands dirty and encounter real problems that tutorials won't prepare you for. Now, let's talk about how to structure your learning time because this matters more than most people might realize. Time blocking is

the practice of scheduling specific chunks of time for specific goals. You

can do this in Google Calendar or whatever app you like the best. When

you're learning a new tech stack, you need uninterrupted focus. This is what it means for me. I'll block out 90inut sessions dedicated purely to learning and building. No Slack, no email, no

and building. No Slack, no email, no context switching. During that time, I'm

context switching. During that time, I'm either writing code, reading documentation, or debugging, nothing else. And why 90 minutes? Research on

else. And why 90 minutes? Research on

focus and productivity suggests that our brains work best in cycles of about 90 minutes before needing a break. It's

long enough to get into the deep work, but short enough that you won't get burned out. Between sessions, I take a

burned out. Between sessions, I take a real break. Walk around, grab coffee, do

real break. Walk around, grab coffee, do something that isn't screen related.

Then I come back fresh for the next block. If you're constantly getting

block. If you're constantly getting interrupted or you're trying to squeeze learning into random 5-minute windows throughout the day, you're fighting an uphill battle. Protect your focus time

uphill battle. Protect your focus time and you'll be shocked at how much faster you progress. Next, let's talk about a

you progress. Next, let's talk about a technique a lot of people have never heard about, but one that's backed by serious cognitive science research, interle. Interleing means mixing

interle. Interleing means mixing different topics or problem types during your practice sessions instead of focusing on one thing exclusively before moving on. So, instead of spending 3

moving on. So, instead of spending 3 hours only on React state management, you might spend time on state management, then switch to rooting, then work on API integration and cycle back through. I want to point out that time

through. I want to point out that time blocking and interle aren't opposites.

So time blocking controls your environment. It protects you from

environment. It protects you from external interruptions. So Slack, email,

external interruptions. So Slack, email, meetings. But interle controls your

meetings. But interle controls your learning strategy in the session. It

changes how you rotate topics within a focused session. I know interle might

focused session. I know interle might sound counterintuitive. Wouldn't it be

sound counterintuitive. Wouldn't it be more efficient to master one thing before moving on? But research shows that interle leads to significantly better long-term retention and better ability to transfer your knowledge to

new situations. A psychological study

new situations. A psychological study found that students who use interle practice nearly doubled their scores on delayed tests compared to those who studied in blocks. The reason is that

interle forces your brain to continually retrieve information and discriminate between topics. So you're not just

between topics. So you're not just memorizing a procedure, you're instead developing actual problem solving skills. When you're learning a tech

skills. When you're learning a tech stack, this means that you shouldn't just grind one topic until it feels comfortable. Mix it up. Work on the back

comfortable. Mix it up. Work on the back end for a bit, then move on to the front end. jump from database queries to

end. jump from database queries to authentication. This approach might feel

authentication. This approach might feel harder at the moment, but that difficulty is exactly what makes the learning stick. Now, let's talk about

learning stick. Now, let's talk about how to use AI as a tutor to speed up your learning progress. AI tools like Chatbt, Claude, and Copilot aren't just code completion tools. They're also

incredibly powerful learning accelerators when you use them right.

Here's how I use AI when learning a new stack. First, I use it to explain

stack. First, I use it to explain concepts at the level I need.

Documentation can be really dense and assume background knowledge you don't already know yet. AI lets you ask follow-up questions until something actually clicks. So, you can say,

actually clicks. So, you can say, "Explain this like I'm 10 years old with analogies." Or, "What's the equivalent

analogies." Or, "What's the equivalent of this in Java?" And I get tailored explanations instantly. Second, I use AI

explanations instantly. Second, I use AI to review and explain code. So, when I'm reading through example code or open source projects, I'll paste snippets into an AI and ask it to walk me through what's happening line by line. This is

way faster than googling every function and piercing things together yourself.

Third, I use AI to help me when I'm stuck, but I don't let it write all my code for me. The temptation is real, but if you're outsourcing all the thinking to AI, you're not actually learning. So,

use it as a tutor, not a crutch. [music]

Let it explain, let it guide, but make sure you're the one actually implementing and understanding. When you

combine AI tutoring with active project-based learning, you've got an incredibly efficient feedback loop. You

build, you get stuck, you get unstuck with AI help, you understand why it works, and you keep building, and so on and so forth. One more strategy that's underrated is to document as you learn.

This forces you to articulate what you've learned in your own words. This

could be a personal notion page, a Google doc, or a read me in your project repo. I've realized over the years that

repo. I've realized over the years that when you try to explain something, you quickly realize where your understanding is fuzzy. It's also incredibly useful

is fuzzy. It's also incredibly useful for your future self. I can't count how many times I've gone back to my own notes months later when I needed a refresher. So, in conclusion, let's

refresher. So, in conclusion, let's recap the fastest ways to learn a new tech stack. First, don't just do

tech stack. First, don't just do tutorials. Build projects immediately

tutorials. Build projects immediately before you feel ready. Second, protect

your focus with dedicated time blocks.

Third, use interleing within those time blocks to mix topics and build real problem solving skills. Fourth, leverage

AI as a personal tutor to accelerate your understanding. And fifth, document

your understanding. And fifth, document what you learn to solidify your own grasp of the knowledge. These strategies

have worked for me across multiple languages, frameworks, and text stacks throughout my career, and I'm confident they will work for you as well. And

that's all I have for you in this video.

If you found this helpful, make sure to hit that like button, hype the video, and subscribe for more software engineering content. Thanks for

engineering content. Thanks for watching, and I'll see you in the next one.