There’s a lot of uncertainty in how AI will impact jobs, and honestly, most of the predictions, good and bad, are probably wrong in the details. But one thing is crystal clear: software engineering as we know it is over. Not the spirit of engineering, but the particular flavor that dominated the industry for the past decade, the kind measured in lines of code shipped, story points burned, and JIRA tickets closed.

The Rote Era of Software Engineering

Over the past decade, companies have spent the majority of their software budgets on engineering teams essentially tasked with cranking out code. The work, while valuable, was largely mechanical in nature: trace through an SDK’s documentation, wire up an API, ensure the infrastructure doesn’t fall over at 2am, build yet another monitoring dashboard, debate which logging library is the most idiomatic choice this year. Security, stability, syntax — these became the primary currencies of the craft.

That’s not a knock on the engineers who did this work. These are real problems that real systems require. But let’s be honest about what most of that work actually was: careful, disciplined pattern-matching. Learning a framework, applying it. Understanding a cloud provider’s best practices, following them. Writing boilerplate that looks slightly different from the boilerplate you wrote last month. The ceiling on this kind of work was always going to be hit by a sufficiently capable language model. And here we are.

AI coding assistants can now handle massive portions of this layer: boilerplate generation, documentation synthesis, security pattern checking, infrastructure-as-code templating. Not perfectly, not without oversight, but well enough to radically compress the human time required. The rote era isn’t winding down — it’s ending.

What Engineering Was Always Supposed to Be

No kid dreams of just staring at code all day. Nobody picked up their first computer thinking, someday I want to argue about tabs versus spaces and write YAML configuration files for a living. They wanted to build things. They wanted to see a real problem solved and people delighted by something new that made their life easier.

The best engineers have always known this. The ones who built tools that genuinely delighted people, who solved problems that seemed intractable, who made someone’s workday easier or their diagnosis faster. They were never really thinking about code. Code was just the best tool available for translating ideas into reality. The goal was always the idea, always the impact.

AI is stripping away the syntactic layer and forcing a reckoning: what problems do you actually care about? What do you want the world to look like? If you can describe the solution, AI can help build much of the scaffolding. What to build, and why, is more valuable than ever.

A Moment to Reconnect With Your Original Mission

If you’ve spent years in software, there’s a good chance you’ve had moments where the work felt genuinely meaningful, and longer stretches where it didn’t. Now is a good time to ask yourself what human and world problems you actually dreamed about solving when you started down this path. The infrastructure concerns, the syntax debates, the framework churn were always obstacles between you and the thing you really wanted to do. AI is lowering those obstacles.

I’ve been thinking about this myself, and it led me to want to build something small but purposeful. One of the things I find genuinely troubling about where we are as a society is the explosion of misinformation and disinformation, amplified and spread by social platforms at a scale we’ve never seen before. At the heart of a lot of this problem is a conceptual confusion most people carry around without realizing it: the difference between cynicism and skepticism.

Cynicism says everything is corrupt, nothing can be trusted, it’s all a lie. It’s actually a kind of intellectual laziness masquerading as sophistication, and it shuts down inquiry rather than opening it. Skepticism, on the other hand, is disciplined and constructive: it asks what’s the evidence, what’s the source, what would change my mind? It’s the foundation of critical thinking and the antidote to manipulation.

So I built a small applet to help people understand this distinction. It gives you a simple, interactive way to calibrate your own thinking and recognize when you’ve slipped from healthy skepticism into unproductive cynicism. Give it a try:

The Engineer’s New Job Description

The shift isn’t just philosophical. It has practical implications for how engineers should be thinking about their skills right now. The engineers who thrive won’t necessarily be the ones who can write the most elegant recursive algorithm or who know memory management inside out. They’ll be the ones who can identify a real problem worth solving, understand the people affected by it, and use AI tools to build a solution faster than anyone thought possible.

That’s not a lesser form of engineering. It’s a harder one in many ways, because problem selection is ruthlessly difficult and taste and judgment can’t be autocompleted. But it’s also the version of the job that actually resembles what drew most of us to this field in the first place.

The code was never the point. It was always just the medium. Now we finally get to focus on the message.