profile

tanaypratap's letters

How to learn a new stack

Published about 1 year ago • 3 min read

The way to learn a new stack is not to learn a new stack unless you have a depth of knowledge in at least one.

Don't just learn a new stack because it's popular or the latest trend. In the past 10 years of my career, I've moved from one tech stack to another, from Python to JavaScript, then Angular, React, and even picked up GraphQL while working at Microsoft.

When you have depth in one tech stack, it becomes easy to learn a new one quickly and efficiently. So how can you build depth?

Understand the "why".

When Python exists, what is the need for JavaScript? If Django is available, why was Node.js created? What's there in Node.js that's not there in PHP? You get the idea.

When you know why a tech stack or framework was created, you understand the design choices of the author. From small libraries to bigger frameworks, a single design framework is followed throughout almost all the time.

To start the process, it is a good idea to watch talks by the creators and authors themselves. You can even look into the talks of senior developers of that stack. As an example, you can watch this podcast where the creator of TypeScript talks about the gap that the language fills.

Talks are a good place to start understanding why a tech stack was created.

When you understand the "why", it gets easier to pick up the stack because you understand why the author made a certain choice or what he thought while building the stack. This process takes practice and won't happen all at once. But it's worth the effort.

Build small apps.

The next step is to build small applications in the language you want to go deep in. A simple to-do app also works. But you need to spend time writing code and pushing it to GitHub.

When I was learning Node.js back in 2015, I did the same. No one ever looks at that repository. But it is a useful time machine that I can go back to.

Building an app in the framework also creates your Proof of Work, or is at least a start in the direction. It brings in motivation that is enough to push through the journey of learning.

Moreover, when you write code, you get familiar with the syntax. The dexterity you gain helps understand the next level of the semantics of the language better.

Write production-level code.

Now you can't learn how to write production-level code through tutorials. The way to do it is by watching the talks of developers at conferences and reading their blogs.

You can create a list of such developers on Twitter and follow and interact with them as well. They might not always talk about the tech stack but they will surely share insights from time to time. And when you interact with them for long enough, you can even directly ask your questions - of course, considering that you have already done your research and tried to find the answers.

Best practices

Most people who know a tech stack rarely know its best practices. And this is where you start to set yourself apart. When you have spent time on all the previous steps, you understand the design choice and you can solve prod-level problems, you become the expert in the team who can educate others.

Find out podcasts where deep discussions about languages, frameworks, tech stacks, etc. happen. Listen to them on 1.5x or 2x when you are walking, in the gym, etc. You will not just build depth and learn the best practices of the language but also become aware of the future of the framework and where it is heading.

All this while, don't forget to take notes just like you used to do in school or college. When you subconsciously spend time learning one framework and follow a methodical approach, you will truly become an expert in the field.

Read source code.

Now that you have covered the four steps (and this is important), you can start reading the source code of the tech stack. It will take time to reach this level where you can read and understand the source code. But this is where you will start deriving real joy and satisfaction from.

Vertical skilling is what builds craftsmanship. When you spend so long on the stack and enjoy the process of learning, you naturally become curious and want to explore how the stack was created. Then you replicate the APIs and write your own, smaller version. You can write your own React, Redux, React router, GraphQL, and whatnot.

But you need to follow the process.

Vertical skilling takes time, so it is important to follow a structured approach. This involves spending time on best practices, tutorials, and apps, and following a set method.

By going deep into a particular technology, you'll unlock new powers that will set you apart and establish your expertise in the field.

Till next time

If you liked the newsletter issue, subscribe and share it with your friends and family. Ensure that you mark the address as important so that you don't miss out on any future issues.


twitterinstagramlinkedinyoutube

tanaypratap's letters

I write about mentorship, education, tech, career, and startups.

Read more from tanaypratap's letters

Do you know what Indian students are best at? Don’t read ahead and try to take a guess. It’s not programming, finance, or marketing. It’s aptitude. Lakhs and crores of students apply and prepare for government exams every year. They all are extremely hardworking and they are amazing at solving aptitude-based questions - questions that you and I would fail to solve. Although the odds of cracking a government job is one in tens of thousands, these students have much that we can learn from....

22 days ago • 3 min read
red and white square illustration

As much as I appreciate YouTube as a platform to truly interact with one’s audience, I strongly feel that it has led to significant harm to students who wish to learn a skill. Why? Simply because of the hellish number of choices, most of which aren’t helpful at all. I have talked to enough freshers and if there’s one thing that I am sure of - it is that all of them are stuck in tutorial hell. what is tutorial hell? Tutorial hell can be described in three stages: The Watch-a-thon​In the...

about 2 months ago • 4 min read
a computer screen with a web page on it

People were still settling in with ChatGPT3 when its updated version ChatGPT4 was released. On top of that, there are rumors of Q* and OpenAI achieving AGI. Then there's Google and Gemini. The side-effect of these advancements in AI is the fear that AI will take away jobs. You do not want another threat in the current market where jobs are already scarce. My YouTube comments are full of them. I believe that the fear, as real as it seems, is quite misplaced. the constant fear of...

2 months ago • 3 min read
Share this post