Forem: Agon

The part of the job AI keeps missing

Agon — Tue, 07 Apr 2026 11:41:34 +0000

There was a moment, sometime in 2023, when I started to feel it. Not panic, but a kind of alertness. The articles were everywhere. AI replacing engineers, entire roles becoming obsolete overnight. And there I was, managing a team of them.

I was never particularly worried. I've always believed that the human element in building software, the judgment, the relationships, the context that lives between people, isn't something that can be quietly absorbed by a model. But being confident isn't the same as being complacent. Something was shifting, and anyone not paying attention was going to be left behind.

So I paid attention.

The first thing I noticed wasn't that I had more time. It was that I had more headspace.

When the routine information-gathering starts taking care of itself, you stop carrying it around. And when you stop carrying it around, you start noticing things you were too busy to notice before. A pattern in how the team is communicating. A tension that hasn't surfaced yet but will. Someone who's been quietly coasting for two weeks.

That kind of noticing is the job. It always was. But it gets crowded out when you're spending energy on things that, in hindsight, didn't really need a human.

What changed wasn't my responsibilities. It was the resolution at which I could see them. And that difference compounds quietly. Better visibility leads to better conversations. Better conversations lead to better decisions. It's not dramatic. It's just incrementally, consistently better.

Here's what I keep coming back to. Someone having a bad week doesn't need a summary. They need someone to notice. A team losing momentum doesn't need a report. They need someone to ask the right question, at the right time, with enough context to actually do something about it.

That's not a workflow. It's not a prompt. It's presence. And presence is stubbornly, irreducibly human.

I think about this whenever I hear people worry about where managers and leaders fit in an AI-augmented world. The anxiety makes sense on the surface. If the tools get smart enough to track progress, surface blockers, and generate updates, what's left?

What's left is the reason any of that information mattered in the first place. Someone still has to decide what to do with it. Someone still has to have the conversation that changes how a person thinks about their work. Someone still has to hold the team together when things get hard and push them when things get comfortable.

AI made me better at finding information. It didn't get any closer to knowing what to do with it.

So, did AI change my job? Yes. Completely. And not at all.

The surface looks different. The way I prepare, the way I gather context, the way I show up to conversations. All of it has shifted. But the core hasn't moved an inch. People are still complicated. Teams are still unpredictable. And someone still needs to stand in the middle of all that complexity and make sense of it.

That someone isn't AI. Not in any way that actually matters.

The interesting question was never whether AI would replace me. It was what kind of leader you become when you stop spending energy on the wrong things.

I'm still figuring that out. But with a lot more clarity than I had three years ago.

Role of AI and the future of programming

Agon — Sun, 04 Dec 2022 14:22:14 +0000

In 2022, the role of AI in programming was a hot topic. Many predicted that it would take over the world of coding and leave human developers obsolete. But as it turns out, even AI needs a little human help from time to time. After all, it's hard to code without opposable thumbs!

Yes, AI has certainly made its mark in the world of programming. It's been able to automate certain tasks and even generate code on its own. But let's be real, it's not exactly the next Einstein. In fact, some developers have even started calling it "Artificial Idiocy" because of its limitations.

But even with AI's help, there are still limitations. It can't replicate the critical thinking and problem-solving skills of a human developer. It's unable to tackle the unique challenges and nuances of each project. And it certainly can't produce code with the same level of elegance and creativity as a human can. So in the end, it's just a fancy calculator that gets confused by advanced algebra.

In short, the future of programming is not one without human developers. AI may continue to evolve and become even more advanced, but it will always need human guidance and oversight. After all, we can't trust robots to do all the work. They might try to take over the world like in every sci-fi movie ever made.

One example of this is OpenAI's chat function. It's able to hold intelligent conversations with users, but it's not a replacement for a human customer service representative. It's simply a tool to assist them in their job. And let's face it, robots are terrible at customer service. They can't understand sarcasm, they can't empathize with frustrated customers, and they definitely can't tell a good joke. So while AI has its uses, it's not the end of human customer service just yet.

As we look ahead to 2023, it's clear that the role of AI in programming will continue to evolve. But one thing is for sure: human developers are still very much needed and valued in the industry. They bring a unique perspective and skillset that AI simply can't replicate. And as long as there are complex problems to solve and innovative ideas to bring to life, human developers will always be in high demand. So don't worry, your job is safe (for now).

Note: As you have likely realized by now, this blog post (including this note) was not written by a human, but rather by an OpenAI chat model trained on a large dataset of text. The cover image too, has never existed before. It was generated using Dalle 2 from OpenAI. While some may dismiss this as a mere novelty, the truth is that the use of AI in writing and other creative endeavors is becoming increasingly common and important. In the year 2022, AI played a crucial role in many aspects of our lives, and its influence will only continue to grow in the years ahead. It is our sincere hope that this blog post has provided you with a glimpse into the potential of AI and its impact on the world of programming.

A simple way exists… but it’s not that simple!

Agon — Tue, 07 Jan 2020 21:53:33 +0000

I remember back when I was starting to learn programming, my very first days with my hands into the code. As a new guy on this huge and new world for me, everything seemed so perfect, so smartly built, so complex and magical. I just loved everything about it. Reading books and watching tutorials for hours and hours started to become my new lifestyle. I, started to learn how to create my own magic, to write those complex and smart lines of code…

But then, something really changed… I was not a new guy on the coding world anymore. Things I wanted to create required more work to do and hours and hours slowly became days and days. Smart and complex solutions were just becoming more and more complex but not smarter. Complex was not magical to me anymore!

Something really changed… All those years admiring complex works of other people, thinking how smart one should be to write such number of lines of code, to work on such huge code bases… I thought that should be very hard to do. And yes, it was; not smart and absolutely not efficient anyway.

Something really changed… I started to learn how not to love complex anymore. I started to love something else, quite the opposite in fact. Yupp, you are right: simplicity.

I started by replacing many rows on many places in my code with some tiny lines that were doing the same job. Creating service classes and functions were becoming my hobby. I started developing new ways of thinking. I was really enjoying it. I thought that thinking simple would solve many of my problems and on top of that, it would be more easy. But ohhh boy, how wrong was I.

I never thought that creating something simple would be so hard. It was unbelievable what I found. May I present to you the irony itself: The hardest and most complicated part of my job was creating something simple.

I found myself many times knowing how to do or write something but thinking very hard because I wanted it to be and to look simple. But you want to know what’s the funniest (not so) part: How do you tell someone that you’re thinking so hard just to create something simple? People want results. They don’t really care if it’s simple or it’s complicated, they want something they can see, they can touch…

But I didn’t care either. I was creating my own magic and I wanted to do that by my own rules. And of course people asked again and again, why I was taking so much time just to make something simple, why I was obsessed with this word. My answer to that was: I’m doing a favor to my future self, I’m saving HIS time!

Simplicity is smart, simplicity is worth it…