By Thomas Ptacek (Fly.io) and The Nilenso Team • June 2025

I've been mulling over these two AI programming articles for a week: Fly.io's combative defense and Nilenso's measured playbook. They are manifestos for competing perspectives on what it means to be a software developer in 2025.

The surface disagreement is about tone. Fly.io dismisses AI skeptics as working "out of spite," while Nilenso acknowledges legitimate concerns. But dig deeper and you find they're arguing about what it means to be a developer when the thing you've built your identity around starts dissolving.

The Broken Woodworking Metaphor

Fly.io's cutting analogy should be scrutinized:

"Do you like fine Japanese woodworking? All hand tools and sashimono joinery? Me too. Do it on your own time."

The message is clear: stop treating enterprise software like your personal art project. We're here to ship features, not craft code cathedrals.

Fair enough. The counterpoint is "What helps the human helps the AI," I.e. Good tests, clear documentation, consistent patterns, simple code structure, and so on. The "craft" that truly matters is creating environments in which both humans and machines can function effectively. That does not make elegance a fetish. That is exactly what engineering is.

As all analogies do, the woodworking metaphor falls apart quickly. Nobody's paying a premium for hand-typed forEach loops. The master carpenter makes one table at a time; I can now generate entire systems before lunch. You can't refactor a chair (I don't think). Code is infinitely malleable, which makes the initial "craftsmanship" less important than we initially thought it was.

The real skill was never about writing "code as poetry." It was about making systems that work, that other people can understand, and that can change over time. AI doesn't destroy that skill. It just makes us say what the craft was.

Kids These Days

There is a throwaway line in the Fly.io piece that stopped me cold:

"Kids today use asynchronous agents—wake up, assign 13 tasks, return to 13 PRs."

I tried it. Not 13, but five. By lunchtime, I had reviewed x3 more code than I had written the previous week.

Is this what the future looks like? Probably. Is it okay? I honestly do not know.

The Nilenso folks say slow down, make small commits, and maintain standards. And of course they're right. However, something is intoxicating about the sheer velocity of "vibe coding". Like the difference between carefully composing an email and just firing off texts. Different kinds of communication...

The Thing Nobody Will Say

Both articles dance around the idea that being good at writing code and typing out elegant solutions is becoming less valuable. I spent years honing my skills in this area. And now? Now I watch an AI do it faster, often better, and definitely without the ego.

Mental Models and Mediocre Code

The philosophical divide is stark. Nilenso states that we are here to "maintain large bodies of well-defined mental models achieving business needs." Inspirational! This is probably true.

Fly.io states that "mediocre code is often fine." This is also true in some cases.

It's not really about standards slipping. It simply acknowledges that AI inverts the cost structure of software. When generation is inexpensive, perfecting each line is a waste of resources. The craft transitions from writing elegant code from scratch to designing systems that achieve business objectives.

Hallucination Nation

Fly.io claims hallucination is "solved" through iterative agents that compile, test, and fix. Nilenso stays silent, emphasizing human review instead.

In reality, the severity of hallucinations depends on the context. Strongly typed languages catch more errors. Sophisticated agents reduce impact. But for critical business logic, you still have to read every line. You're still responsible. The agent is just very fast fingers.

Where This Leaves Us

I don't have a clean ending here. Neither do these articles, really. Fly.io wants me to embrace the rocket fuel. Nilenso wants me to maintain my standards.

The hierarchy of software skills is inverting:

1. System design (irreplaceable and increasingly valuable)
2. Code quality judgment (valuable but increasingly automated)
3. Code writing ability (rapidly commoditizing)

Both articles agree that being excellent at writing code is becoming akin to being an excellent typist in the age of dictation. The new differentiator is orchestrating AI while maintaining judgment about what to build and whether it's built well.

The craft isn't dying, it's mutating. I'm not sure what it's becoming. But fighting it is like being a typist complaining about word processors.

The developers who thrive will be those who can manage AI agents like a conductor manages an orchestra, maintaining standards while embracing productivity gains that would have seemed impossible just two years ago.

Once you get over the existential crisis of watching a machine write better code than you, it's exciting. I spent yesterday designing systems instead of debugging semicolons.

And at least they can't write good blog posts yet...