No Going Back
Contents
I'm dictating this post through a Swift app I built over the last couple of weeks. Push a key, talk, post appears. I didn't write the code character by character. I described what I wanted and an AI wrote it.
I don't think we're coming back from this.
The back story
I'm 41 now (that's wild in itself). I've been building websites and apps since I was 13. My first paid computer job was at 16. £2.50 an hour to run 'remove' commands that deleted leftover files from a typesetting system.
Looking back, that job should have been a cron job.
I was literally paid to type commands that could have been automated in five minutes.
One Saturday morning, slightly hungover, I typed rm and got distracted. Came back, typed rm again, then the filename.
Hit enter: rm rm filename.txt
The first rm deleted the second rm. The command deleted itself.
I typed rm anotherfile.txt
Command not found.
Uhoh.
I briefly stared at the cursor wondering what to do. I had to go tell my boss I'd removed his custom 'remove' command. Fair to say the guy was unhappy. He said he didn't have a backup of the custom rm command. Oops.
A script that could delete those files would not have been so unreliable. Stupid human.
Since then: started freelance web design at 19. Music technology at uni while freelancing. SMEs, startups, agencies, startups again. I built Java apps for Google Glass in 2013. Wrote APIs, built front ends, created server rendering pipelines before Next.js existed. I worked on a whitelabel app platform for six years. I wrote a 12-week bootcamp curriculum that's now trained thousands of developers. Built a social network for music. I've been writing code my entire adult life.
I'm certain the game has changed. Writing code by hand is a dying art and the sooner we accept that the better it will be for our future.
The wall
For the last few years, AI has been useful but limited. Early on Copilot changed the game for things I was doing at the time like writing tutorials. I could explain a concept and it would write the example code. 95% accurate, maybe higher. It was impressive but not very disruptive.
Building anything substantial there was always a wall. Some blocker that killed progress. Eventually you'd fall into a loop where fixing one bug would create another, and fixing that one recreated the first. You'd bounce back and forth, never quite solving the whole problem. That was the ceiling.
Opus 4.5, plus some methodical process, broke through it.
The cost
I started on the simple $20 Pro plan. Then I was bouncing between Claude and GPT. Then Claude Code arrived. It escalated to the 5x plan. Now I pay for the 20x plan and I'm creating more software than I've ever been able to create in my life.
It sounds expensive until you see the output. 30 days to build something that would have taken previous teams I've worked with months. And not a rough prototype. It's good software with more polish than I would have ever added manually for an MVP.
What the job becomes
Nobody is going to pay me to type slowly on a keyboard when they can generate thousands of lines of code in minutes. That's just the reality. It doesn't matter how I feel about it.
So what's left? Ideas. Planning. Spec writing. Marketing. Knowing when to nudge the AI in the right direction and when to let it run. There is a lot of technique here, and knowing the technology still matters. Understanding the trade offs and making the right decisions still matters. There's new territory to explore, new ways of working with something that didn't exist two years ago.
I don't necessarily like this fact that things are rebalancing so quickly. I've spent 25 years honing the craft. But liking it isn't the point. It's still true.
The revelation
The code itself isn't the most interesting part. The interesting part becomes the documentation.
A good process can generate requirements documents, flesh out specs, identify risks and pitfalls, create instructional plans that reduce errors and hallucinations. We can keep logs of where a plan deviates or runs into issues, we can review and write verification reports according to the requirements we started with. Every problem encountered, every solution found, every decision made and why. It's all captured.
If you put this corpus into a semantic search, you'd have the entire meaning of the project searchable. Every issue, every fix, every reasoning chain. That level of documentation would have been impossible even with the most diligent team in my experience. The overhead would have been too high. Now it's nearly free.
With AI, context is everything. The documentation that enables the code to be produced is starting to look more valuable than the code.
The future
For now, the models that can produce quality are slow. The models that are fast can't quite match the quality yet. But both sides are improving. We're watching the gap close in real time.
I don't know exactly what software development looks like in five years. But I know it doesn't look like me typing on a keyboard character by character.
We're not coming back from this.