Somewhere in the last year, a bunch of us who hadn’t shipped real code in years quietly started shipping again. Not reviewing it. Not architecting it in a doc and handing it off. Actually building it. And loving it.
I lead software for a living. That means my days are meetings, roadmaps, hiring, unblocking people, and saying “no” a lot. The last time “writing code” was my actual job was a long time ago. I never stopped being able to code — I stopped having the time to code. At least in my day job. There’s a difference, and every engineering leader reading this feels it in their bones.
If you identify with this post and you are in a manager position, please heed my warning. Throw away your prototypes. Stop vibing “quick fixes to unblock me”. Stop trying to help your devs because you think you’re superman now. You are only pissing everyone off, I promise. The devs who didn’t stop coding while you were off doing important manager/product stuff are the ones who understand what’s best for the codebase. And yes, they also know how to use LLMs.
Just because you feel like you have a new superpower doesn’t mean that everyone needs to be saved. Follow the damn engineering process.
100%
I got this my superior who decided AI allowed them to code fixes and we were stuck with buggy code that wasn’t merged, because the superior had moved on to a more important task requiring more AI “support” and couldn’t process the comments. Things could’ve been faster had we implemented things ourselves.
When you aren’t out in the fields dealing with stuff, you don’t know what’s important to the workers. “Jumping in” for a minute to do a task might actually slow down the team if understanding of day to day struggles isn’t present.
I’m contributing story execution. And my goal is to contribute like any other team member. My code gets reviewed. Harshly because I don’t know all the norms for this code base. And honestly it’s great because to be a technical lead I feel like I should know the code way better than I’ve had time to.
Great. The “why don’t you just…” idiots are “writing” code again and not reviewing it.
What could go wrong?
Yeah, I thought the blog post was gonna be about horror stories, because even the greatest senior/architect/whatever will push absolute garbage into your codebase, if they don’t have the time to familiarize themselves with how the code actually works.
Even if it’s some isolated component, if they implement it without explaining the what and why to others, then no one else can maintain or debug that. And you do not want to rely on your super important guy to have the time to fix things when production grinds to a halt.
At this point I think we’re moving past maintaining projects. Tech debt is no longer going to be paid off; management will just suggest a complete rewrite using AI as soon as issues get too complex for a junior dev with Copilot to figure out. It’s going to be a mess.
Who needs human review, the AI can do that? Who needs human coding, the AI can do that? Who needs human testing, the AI can do that? Who needs human requirements engineering, the AI can do that.
80% uptime. How did this happen?
Still five nines of uptime — 79.9999%.
The AI agents need to sleep, too, so the downtime is expected
Could you qualify what you mean by “idiots”? The author and the examples he gives - Kent Beck, David Heinemeier Hansson, and Garry Tan - are all very experienced software engineers.
By idiots I mean the people who haven’t engineered in years.
The subject of the article.
Seems self-evident.
Well that depends on how quickly you believe that engineering skills atrophy without use.
Though, I’d argue programming is like riding a bike. Even after a long break, it’s not difficult to get back into it, as the fundamentals for programming haven’t changed.
If they’ve stopped engineering, then they probably weren’t good in the first place, what skills they had are rusty, and their knowledge is out of date.
I’d argue programming languages are like natural languages. Use it or lose it.
I don’t mind writing code and having copilot review it. I can tell a clanker to fuck off in satisfying ways that I couldn’t with a junior dev.
…and there’s the one time in 10 when it sees something and I can make a minor improvement.
Still, the “Oh crap, my code actually is broken” monents are few and far between.
An agent handed to someone who doesn’t know what they want produces a confident pile of plausible nonsense. Fast. The same agent handed to someone who knows exactly the shape of the solution and can smell when it’s drifting? That’s the mech suit. That’s the leverage.
yesssssss
Agreed. This is my experience exactly, though I really only took a break from production code for about a year. It feels really good.
Writing code was never the bottleneck. This didnt make it easier. It just made it faster, people already ignored the output when engineers made it.
Reminds me of that one time when Microsoft said “We don’t need UI/UX designers, powerpoint is so easy that managers can do UI/UX design themselves”.
That’s how we got Windows 8.
Weather or not this is good or not wildly depends on who the person is and what the situation is. The article seems rather self aware.
Not reviewing it. Not architecting it in a doc and handing it off. Actually building it.
??? Reviewing, architecting, and handing it off is pretty much what AI coding is. I feel like the entire industry has some kind of AI psychosis or something.
I mean, building again helps leaders stay grounded.






