I always worry about keeping myself updated in this fast-paced era. But I've noticed that I stay quite up to date compared to a good number of developers around me.
I want to share my journey with the people around me, especially new engineers entering this fast moving bullet train called software engineering.
In 2023, when I was very new to university and only knew a bit of competitive programming and had just an introduction to building a basic web app using JavaScript, luckily or maybe being good at CP, I got an internship at a fintech organisation in Bangalore, India. Even more luckily, I got a project involving RAG, LLMs, and search algorithms. That pushed me into understanding how data science and AI are shaping the world.
In 2023, ChatGPT was still relatively new, but I sensed the future (or at least what is happening today) and how it would shape developers' lifestyles and the future of engineering.
I was a bit scared, but more importantly, I got an opportunity to learn more and stay hungry to grasp everything happening in the software world. I will share a roadmap and a few tips so that you can stay sharp in this fast-paced future.
FYI, I am a developer who rarely works on core ML or LLMs, but I am still knowledgeable enough to understand new papers published by big players like DeepSeek and Meta. I am aware of the potential of the latest models, frameworks, and developments.
My day-to-day work does not involve ML or AI (except some vibe coding using Claude and GPT). Rather, it involves distributed systems, backend engineering, infrastructure, and related areas. I will guide you on how you can stay in the race.
I am a mechanical engineer by formal education. So most of my university courses were not focused on the foundations of ML, AI, or the mathematics required to understand modern AI systems and deep learning.
I remember I got a free Udemy course (which is currently not free). By the way, I have never spent money on a Udemy course except one that was not related to software. That course covered different machine learning algorithms and techniques.
I did not code along with it, but I made sure I watched almost every video and understood the concepts behind each algorithm. That built my fundamentals for the future. From there, I never looked back.
Here are the resources I recommend each of you follow to build the basics if you are not a data scientist or ML engineer but still want to stay updated.
https://www.youtube.com/@codebasics/playlists Watch an ML tutorial playlist and a DL tutorial playlist from this YouTube channel. You can code along if you want. Understanding the basics of all algorithms is a must. Do not skip the fundamentals. Building a habit of understanding fundamentals will help you keep up.
You can also cover RL or image processing if you are interested.
I always recommend building the habit of grasping concepts after reading about them and making sense of them. This will help you later when understanding research papers in topics you are interested in.
Read Clean Code by Uncle Bob. That is a must. It can push you a year ahead in experience.
Read at least one book on architecture. Pick Clean Architecture by Uncle Bob or even Domain-Driven Design by Eric Evans. At least one is fine.
You can visit www.ujjwalraj.com to get more reviews on what I have read and recommend to beginners.
If you become more interested in this area, you will automatically start picking up more books, courses, and learning resources in this domain. If not, someone else will pick this path, and you will become an expert in some other domain.
If you are a working professional, you should understand everything in and around what you work on.
I got this advice from the CTO of a very popular startup in Bangalore. Trust me, it is worth following.
For example, I know how Python works behind the scenes. I know how Docker's architecture works. I understand Kubernetes. I literally read a book on Docker and another on Kubernetes to satisfy my curiosity. That said, going through a good YouTube video is also fine if you want quicker results. But be confident about it. Confidence gives you more energy and hunger to learn.
People recommend reading newsletters. But I find it very difficult to maintain the discipline consistently. Eventually, it can exhaust your hunger to learn more.
So instead, join a good community that has discipline, or maybe a cohort (preferably free).
If that's not possible, build one. I actively participate in a software reading club at my company, which I joined in the second hour after joining the firm. A link to the group appeared on our public Teams channels, and I joined immediately. I'm glad I did.
We have weekly reading sessions, and I love volunteering as a moderator. We covered three books last year by spending just one hour together every Friday.
This is also important to keep the spark alive.
Participate in online or offline hackathons with challenging problem statements if you have the time and bandwidth. The agenda is - Build Things, Don't Just Consume Content.
I am recommending this in 2026. This is a must because Claude is shining right now. Later it could be OpenAI, Gemini, or someone else—we never know.
The goal is to keep yourself up to date with AI-assisted coding tools so you can be more productive, create more time for yourself, and ultimately spend more time on life outside work.
This is not mandatory, but if you want to understand papers released by DeepSeek, Meta, and others, it helps a lot—especially if you are not from an ML or data science background and are more focused on software development.
I use Todoist, by the way, and have organized it like a Jira board. It's up to you how you maintain a To-Read list. I would love to hear recommendations on this as well.
Maintaining a To-Read or To-Learn list helps you track your progress. I have a lot in my backlog, and this is a glimpse of it.
You don't have to read everything in one day or one week. But you do need to be consistent. So you need to track everything.
Add all your hot topics there and start picking them up when you are bored or when you feel like learning something during a dull weekend.
Once you are here, don't stop. Keep doing it because the world is moving fast.
Learn time management. I think engineers are naturally good at it.
Once you become good at learning quickly—which will take around 3–4 months if you follow the advice above—you are going to save a lot of time.
Be patient and learn how to learn fast. The steps above provide a good roadmap if you don't know what to learn or where to start.
In the last year, I got premature-promotion at work, learned swimming, started volunteering more in the reading club, started working out 3–4 times a week, and wrote technical blogs as well. Recently I started learning to sketch (the cover image of this blog is sketched by me).
I attended family functions in my hometown and all major festivals. I am very busy at work, but I still have a lot of time left for myself, my friends, and my family.
Here, I am not lucky—I am developing this skill.
I used to struggle a lot, but I learned this from a senior architect at my previous company, where I joined a cohort. (That's why I recommend joining a good community.)
I still need to improve a lot, but I am definitely better than I was a year ago.
Earlier, I used to get nervous if I failed to learn anything in a week.
But slowly, I realized that this is fine.
It is normal to have an unproductive week, but don't forget to add things to your backlog or To-Do list; otherwise, you will forget them.
One piece of advice I would give is this: always keep your brain under some form of productive stress.
Software engineering is not only about knowledge; it is also about thinking. If you stop challenging yourself for long periods, you may notice that your problem-solving speed, curiosity, and ability to grasp new concepts start to decline.
You do not have to participate in coding contests if that is not your thing. But you should regularly do something that pushes your limits.
That could be:
If you do not use your muscles, they weaken over time. The same applies to your ability to learn and think deeply. Keep your brain under load, and learning new things will become significantly easier throughout your career.
If you take only one thing from this article, let it be this: You do not need to be the smartest engineer in the room. You do not need to read every paper. You do not need to know every framework. You only need to remain curious and consistent. Technology changes quickly, but the ability to learn has remained the most valuable skill throughout every generation of software engineering. Build that skill, and you will adapt to whatever comes next. I would love to hear your thoughts and how you keep yourself updated in this rapidly changing industry.