Marian Sorca on Why AI Is the Best Thing That Ever Happened to Programmers

It Started With a Post on X

I saw a post on X the other day from a programmer saying AI coding took the joy out of programming.

My reply was simple. If you want to code the traditional way, you still can. On your own time, side projects, open source, personal tools, nobody is stopping you. If your employer is enforcing AI-assisted workflows, that is a different conversation. But the craft itself is not going anywhere.

There will probably always be a small group of people and companies who code fully by hand. That is fine. I have respect for it.

But the more I sat with that post, the more I realized the whole debate is missing the point.

Business Never Wanted to Write Code

As a business owner, my goal is to benefit from software. Not to write it.

Think about a local pizza place when the web arrived in the late nineties. Their goal was never to hire developers and build a custom site from scratch. Their goal was to be findable online, show their menu, and take orders. The website was the means. The point was more pizzas sold.

Same story with the dentist who got a booking system. Same story with the accountant who moved from paper ledgers to QuickBooks. Same story with the freelancer who set up a Stripe link instead of chasing invoices.

Nobody woke up wanting software. They wanted outcomes. Software was the cheapest path to those outcomes at the time.

The goal of business has always been the same: benefit from software. Internet, then web, then apps, now AI. The layer keeps changing. The intent does not.

The Ground Was Stable for Fifty Years

Here is the part that makes this moment genuinely different.

For the last fifty years, the way we produced that software stayed remarkably stable. You wrote code by hand. You learned a language, then another, then a framework, and the skills compounded. A high schooler could map out a career path in tech and trust the ground would not move under them.

That ground is moving now. And I want to be honest about what that feels like from the inside.

As a software engineer with more than ten years in, I get why this feels disruptive. The day-to-day has shifted from writing code to planning, guiding, supervising, editing, and testing code. That is a real change. It is fair to grieve parts of it. The person who loved being heads-down in a single function for three hours, just them and the problem, has lost something real.

I am not going to pretend otherwise.

But This Is Not Only a Software Story

Every kind of digital work is in the same shift right now.

Writers are using AI. Designers are using AI. Analysts, marketers, lawyers, accountants, customer support teams, researchers. The work is not disappearing. It is being restructured around a new layer of leverage.

This is a historical moment. The productivity gains on the table are enormous for anyone willing to adapt. And for all of us as consumers, the next five years will bring tools and services that were simply not buildable before.

Software engineers are not uniquely affected. They are uniquely positioned.

The Reframe I Want to Offer

If a non-technical person can now ship a working product with AI, imagine what a programmer can do with the same tools.

Think about what you actually know. You know how systems are structured. You know what good architecture looks like, and what falls apart under load. You know how to debug something when the AI is confidently wrong. You know which corners can be cut and which cannot. You know the difference between "it works on my machine" and "it works in production."

That knowledge did not disappear. It just became more valuable, because now there is a tool that can execute on it at 10x to 20x the speed of hand-written code.

The gap between "I have an idea" and "I have a business" has never been this short. And the people best equipped to close that gap are the ones who already know how software works.

Two Questions, Not One

So the question worth sitting with is not "is AI killing programming?"

It is two questions.

How do I use my coding knowledge plus modern AI tools to build something of my own?

Maybe that is a SaaS. Maybe it is an internal tool you turn into a product. Maybe it is a freelance service stack that used to require three people. The opportunity surface is wider than it has ever been, and it is not closing anytime soon.

Or, how do I use these tools to make the job itself more interesting again?

If you stay employed, you can push for the harder problems. The ones where AI accelerates the grunt work and you spend your time on architecture, edge cases, and the creative parts of the craft that made you fall in love with it in the first place.

Both answers are available. You just have to pick one.

The Part Most Engineers Already Know

I left a comment on my own LinkedIn post that I want to repeat here.

Remember, programmers, we are the ones who like to try and use the latest and greatest technology first.

That is literally the job description. The whole culture is built on it. The person who refused to use IDEs because vim was sacred, the one who refused to use frameworks because real engineers write everything themselves, the one who refused to use the cloud because bare metal is purer. History is full of those people. They are always sincere, and they are always bypassed.

If there is one group of people who should be leading this shift, experimenting first, shipping faster than everyone else, and redefining what software production looks like, it is programmers.

We get to go first. That is the privilege of the job.

The Takeaway

AI is not the end of programming. It is the moment where programmers stop being the bottleneck in their own ambitions.

The business owner in me is genuinely thrilled. I can ship things now that would have required a full team two years ago. The engineer in me is even more thrilled, because the gap between what I can imagine and what I can build has never been smaller.

If you are a programmer feeling stuck in the debate about whether AI killed the craft, I would offer this. The craft is still there. The tools got better. The ceiling went up. And the people who move first are going to own the next decade of software.

Pick one of the two answers. Build something, or make the work more interesting. Either way, do not sit this out.