The Lost Feeling of Building

I honestly don't know where to start from.

It all started when I wrote my first line of code back in December 2023. Back then, everything felt pretty simple. Coding still looked exciting to me. I was learning best practices, trying to understand how things worked, watching tutorials for hours, and getting excited over the smallest things. Even fixing a tiny bug felt rewarding because it felt like I was actually learning something.

At that point, programming still felt creative. It felt like building.

Then 2024 happened.

I interned at a company, and around that same time, AI tools, especially ChatGPT, started becoming part of everyday development. Everyone was talking about it. It was trending everywhere. Inside the company, we were basically encouraged to resolve our doubts and queries through ChatGPT because it was fast and efficient.

And honestly, it worked.

You would copy an error, paste it into ChatGPT, get a fix in seconds, throw it into VS Code, and move on with your life. That slowly became the workflow. At first, it felt magical. It genuinely felt like software engineering had entered a completely different era.

But over time, I started realizing something.

I am what you would call a post AI learner. I started learning programming in a world where AI already existed and was available twenty four seven at my fingertips. That changes everything.

A lot of older developers talk about spending days debugging issues, reading documentation for hours, building projects from scratch, and slowly developing intuition about systems and code. They talk about the satisfaction of finally understanding something after struggling with it for days.

I honestly never experienced that process properly.

Every time I got stuck, AI was there. Every time something became difficult, there was always an easier shortcut available. And while that sounds amazing on paper, I think it quietly changes the way you learn.

A lot of things I genuinely wanted to figure out on my own, I never really did. Not because I was lazy, but because the environment itself pushes you toward speed instead of understanding. The goal is no longer to deeply understand a problem. The goal is to solve it as fast as possible and move on to the next thing.

And the scary part is that after a while, you stop noticing it.

You stop asking, "Why does this work?"
You stop asking, "How was this solution even generated?"
You stop exploring.

You just accept the output and continue.

That's where my relationship with AI started becoming weird.

Because on one side, AI is genuinely incredible. There is no denying that. You can build things faster than ever before. A single person can now do the work of an entire team from a few years ago. Tasks that would have taken days can now be done in minutes.

But on the other side, I think we are slowly sacrificing understanding for speed.

Most of the time, we don't really know what the generated code is doing. We don't know why AI chose a certain approach. We don't know the tradeoffs, the architecture decisions, the side effects, or what impact the code can have on the overall codebase.

We are just blindly moving with the flow.

And honestly, even when the AI generated code is technically correct, the problem still exists. Because correctness alone is not enough in software engineering. Understanding matters.

At some point, this becomes extremely painful.

Especially when you have to debug something complicated. Or when a feature breaks unexpectedly. Or when you need to scale something. Or when you revisit code written two months ago and realize you barely understand any of it because half of it was generated through prompts during a rush.

That's when the cracks start showing.

And I genuinely think people who started coding post AI are going to struggle with this a lot more in the future.

Sometimes I feel like elite software engineering, the kind built on deep understanding, strong fundamentals, and craftsmanship, is slowly dying. Not completely, but slowly fading away behind layers of generated abstraction.

And what replaces it might just be endless amounts of AI slop.

Code that works temporarily.
Code nobody understands.
Code written fast, shipped fast, and forgotten fast.

I personally work at a startup, and the pace honestly feels unreal sometimes. I push an insane number of commits every day. Sometimes close to a hundred. Everything moves extremely fast. Features keep coming. Bugs keep coming. Deadlines keep coming.

And somewhere in between all of that, I slowly started feeling disconnected from the work itself.

A lot of the time, I don't even fully know what I'm building anymore. You open your laptop, write prompts, generate code, patch things together, commit changes, attend meetings, and repeat the cycle again the next day.

And after a while, life starts feeling strange.

Because the feeling that originally made programming exciting slowly disappears.

That adrenaline rush of solving a genuinely difficult problem?
Gone.

That satisfaction of finally understanding a complicated concept after struggling with it?
Gone.

That late night excitement of building something and feeling proud of it?
Gone.

Everything starts feeling mechanical.

And honestly, that's the part nobody really talks about when discussing AI in software engineering.

People always discuss productivity.
People discuss speed.
People discuss replacing jobs.
People discuss billion dollar valuations.

But almost nobody talks about what this does psychologically to developers.

Because at some point, you stop feeling like an engineer and start feeling like an operator.

You log in.
You write prompts.
You copy outputs.
You move tickets.
You commit code.
You repeat.

And slowly, software engineering starts feeling less like problem solving and more like data entry.

That feeling is honestly hard to explain unless you're living through it yourself.

It feels like your brain is becoming passive.

You are technically producing more output than ever before, but internally, it feels like you are learning less and less. Your curiosity starts dying. Your patience becomes weaker. Your ability to sit with a hard problem decreases because your brain knows there's always an instant shortcut available.

And the worst part is that the industry rewards this behavior.

Nobody cares how deeply you understood something.
People care about shipping speed.
People care about output.
People care about delivery timelines.

So naturally, everyone adapts to survive.

And honestly, I don't even blame developers for this. The entire ecosystem is moving in this direction.

I look around at the people I work with, and almost everyone feels exhausted. Everyone has a hundred times more workload now because AI increased expectations for productivity. Since tasks can theoretically be done faster, companies now expect more work in less time.

But humans don't work like machines.

Just because code is generated faster does not mean the mental load disappears. In fact, sometimes the mental exhaustion becomes even worse because now you are constantly context switching, constantly reviewing generated outputs, constantly trying to keep up with systems moving faster than your own ability to process them.

And somewhere inside all this chaos, the joy disappears.

The happiness disappears.
The learning disappears.
The sense of mastery disappears.

You just feel stuck.

Lost.

Disconnected from the thing you originally loved.

And honestly, I think that's the reason I've started questioning what I actually want from my career.

Maybe I want to move into AI engineering itself and understand these systems deeply instead of just consuming them blindly.
Maybe I want to move toward low level systems engineering where fundamentals still matter heavily.
Maybe I want to build things slower again.
Maybe I just want to think properly again.

I honestly don't fully know yet.

What I do know is that AI itself is not evil. I'm not anti AI at all. I even used AI to help correct the grammar in this article. AI is genuinely useful in many ways, and pretending otherwise would just be dishonest.

But I also think using AI for absolutely everything comes with consequences that we still don't fully understand yet.

And I don't think the industry is slowing down long enough to even think about those consequences properly because right now, everyone is chasing growth, funding, valuation, and speed.

Maybe this is the future of software engineering.
Maybe this is just a temporary phase.
Maybe we'll eventually find balance.

I honestly don't know.

But right now, I just know that I feel burnt out, disconnected, and mentally exhausted.

So I'm taking a break for a month.

Not because I hate programming.
Not because I want to quit engineering.

But because I want to figure out what kind of engineer I actually want to become before I completely lose the feeling that made me fall in love with building things in the first place.