This doesn't happen at all for using agentic coding: What the programmer wants and what the boss wants are pretty well aligned. There are corner cases where someone isn't allowed to use LLMs, but does it anyway, but in most cases, the organization agrees.
Unless the teacher's role is to scaffold and support the students in acquiring what the students want, gain trust and lower the disconnect.
You also wrongly assume that requirements can always easily expressed as natural language.
Another point: Software Engineering always starts where tooling capabilities stop. You don't get a competitive advantage by building without engineers what anybody everybody else can build without engineers.
What I think will happen is AI will write code and it will do the best it can to mitigate mistakes prior to rollout, but once rollout time occurs, rollout will be incremental and it will self monitor by defining success conditions at rollout time. The nature of the code will mitigate "catastrophe" to a small group at worst, but most likely initial rollout will just run new versions of the code in a simulated context (language design could benefit from this) and analyze potential outcomes without affecting current functionality.
But when the code goes live... it will be slowly scope changes progressively (think feature/experiment flags) and if it fails in the initial cohort, it will redirect. If success is positive, it will increase the rollout cohort.
This is a normal software engineering practice today, but it's labor and process intensive when driven by humans. But in a world where humans are less involved, this process is scalable.
Counter points to my own arguments:
1. We don't know yet in detail what AI is good at.
2. AI doesn't need to be perfect, just "good enough", whatever that means for a specific project. More failures while saving hundreds of thousands dollars each year might be acceptable, for example.
This I think is the unexplored aspect of what's happening right now. Guardrails around "good enough" systems is where the future value lies. In the future code will never be as good as when the artisans were writing it, but if you have an automated process to validate/verify mediocre code (and kick it back to AI for refinement when it fails) before it's fully productionized, then you have a pathway to scaling agentic coding.
And this could easily apply to every change we made by hand before AI, it was just a tedious process to layer these things into code when we were just fixing bugs and whatnot. In an AI writes all the code world adding this kind of stuff as table stakes for a changeset is zero cost, effort wise.
To me the trend seems to be that AI produce the same challenges as human did before and that the same solutions are helping. Without a good maintainable code base, AI will eventually fail to even fulfill quantifiable requirements of changes.
That's kind of the point of software since the beginning. Nobody cares about the easy stuff that can be produced without much effort and what's possible without much effort has changed dramatically over the years.
This sounds a lot like allowing an LLM to define tests as well as implementation, and allowing the LLM to update the tests to make the code pass. Recently people have come to understand (again?) that testing and evaluation works better outside of the sandbox.
I'd note here that the long arc of software engineering has been commodifying the discipline into tooling. Ask any unix greybeard how shitty modern abstractions are and they'll give you all you can stomach and yet the wheel turns despite their treasured insights.
I think you’re overly hyped if your actually believe this is going to be a reality in 5-10 years.
People were burying their heads.
Today, there are not many of those people left. Some, but not a lot. Because you can only deny reality for so long.
I don't know what the coding world is going to look like in 5-10 years, but everything has changed radically in the space of a year from maybe 10% of people using agents to code to probably 95% of people now. In about a YEAR.
I don't know, but my assumption is these things will get better to a point where they will be automating close to 100% of coding, and deploying, and verifying, etc. The old job we had will be completely changed well before 10 years. I still think us "engineers" will have a role to play, but I genuinely don't know what it will look like.
Last I saw about a week ago, the stats were about 35%. There may be some confusion around this:
1. The absolute number could have remained the same but the sheer volume of vibe-coders who never coded before raised the percentage. For example, if 100 out of a population of 1000 people uses AI then the percentage is 10%. If, over the next year 9k new vibers were created but none of the existing 1000 people changed their workflow, you will see 9100 people out of 10000 people using AI - that's now a 91% rate of people using AI to code even though none of the people since last year changed the way they work.
2. Last I checked, pre-AI, there were about 12m working developers in the world (SO survey extrapolated). As of February this year, CC, by itself, had 60k subscribers. Even if we err on the side of optimism and assume every single subscriber is running the agent, that's still not 95% of developers.
> I still think us "engineers" will have a role to play, but I genuinely don't know what it will look like.
??? We already know what it looks like - "Business Analyst" has bee a role since forever (at least since 1995, when I entered the workforce). If you wanted a role where you wrote no code but merely drew up specs for the programmers to code, you could have had it as a BA.
It's just that few of us wanted to do that as it paid half what an engineer made. Now with the supply of BAs potentially doubling, it will pay a quarter of what an engineer used to make.
And I know this is the same in peers workplaces.
Why would that matter? I'm not giving you my experience of developers next to me, I'm telling you what I gleaned from published reports.
In any case, I know how these arguments go. It doesn't really matter.
No report is going to help because there are no actual figures for token providers at the moment. We'll have to wait for them to IPO before we'll know for sure.
the bar to "start" is lower and the bar to actually competency is higher now, juniors who want to actually learn instead of just pressing enter over and over again will do so regardless of whatever you do to "help" them.
There's no better time to have a curious mind
It’s not all that different than writing code directly and having it turn into a mess they can’t debug—something we all did when we were learning to program.
It is in many ways far easier to write robust, modular, and secure software with agents than by hand, because it’s now so easy to refactor and write extensive tests. There is nothing magical about coding by hand that makes it the only way to learn the principles of software design. You can learn through working with agents too.
Don’t forget to mention that.
This pre-supposes the idea that the business is _willing_ to let that happen, which is increasingly unlikely. The current, widespread attitude amongst stakeholders is “who cares, get the model to fix it and move on”.
At least, when we wrote code by hand, needing to fix things by hand was a forcing function: one that now, from the business perspective, no longer exists.
Now there actually is time to make things robust if you learn how to do it.
What makes you think you are going to be given time to polish it? You would be pushed to another project. You have more responsibilities with none of the growth.
Or you’re getting the model to do the polishing, thereby developing no skills of your own, and we’re back to the start.
The argument seems to be that AI is causing managers to demand faster results, and so everything has to be a one-shotted mess of slop that just barely works. My point is that it doesn’t take much longer to build something solid instead. Implementation time and quality/robustness are not tightly coupled in the way they used to be.
And of course in the current workplace where there’s often a push from managers to use LLMs as much as possible and to put as much work as possible on yourself, in this churn junior will not get to learn anything besides prompting and simple tooling.
None of this requires coding by hand. I can do those things better and faster with agents helping me. That incudes unfamiliar areas where I am effectively a junior.
Y'all need to stop worrying about the kids.
They're smarter than us and will run circles around us.
They're going to look at us like dinosaurs and they're going to solve problems of scale and scope 10x or more than what we ever did.
Hate to "old man yells at cloud" this, but so many people are falling into the trap because of personal biases.
While the fear that "smartphones might make kids less computer literate" is true, that's because PCs are not as necessary as they once were. The kids that turn into engineers are fine and are every bit as capable.
Thinking is happening at a higher level. Humans are adepts at abstraction, and they are always capable of looking under the hood when needed.
We've never had so much societal capability as now. And that's only going to accelerate. Smart people will use these tools effectively. Don't be so bearish on human ingenuity.
Think of these tools as bullshit / busy work removers. You can focus on what matters and get more done than ever before. Deeper work, more connective work. It also opens fields of research up in an interdisciplinary fashion. People might explore outside of their limited domain now that they have help.
The earth will move on. Smart people will continue to exist. I'll bet on the smart people.
It's an impressive force multiplier, but there has to be some force to multiply.
For example preliterate people have absolutely insane memory. In comparison my memory sucks. Having to use notes, look things up etc sucks. Literacy is a tradeoff but at least it can be argued to be worth it.
Then there is smartphones. This is not the same. The tradeoffs compared to pre smartphones cannot be argued to be worth it imo and I was 20 years old when they were introduced. They make society and lives worse. It's not just about not being able to use PC but your attention and social skills sucking.
Then there is AI which is even worse than smartphones. The tradeoffs are so unthinkably bad I can't really even describe it.
The proof is in the fact that the savvy Atherton dwellers work hard to keep their kids away from the crack they themselves have foisted upon the world, or at least to delay or forestall the encounter.
Due to English-language limitation my most adult life, I struggled to code. Used visual coding etc. But of course, I can't make a living on drag-and-drop harness.
Comes in GPT-3.5, accelerated my learning. Now I'm running my incorporated company, just launched one software-hardware hybrid product. Second one is a micro-SaaS in closed beta.
The point is: when people use "juniors" as a fixed shaped blobs of matter, they focus on the juniors that were in any case going to make mistakes: AI or not. Misses the key point of agentic usage.
> accelerated what learning? learning to code? learning to engineer? learning to manage? learning to market?
I'm pretty certain that you think you're talking to an owner of a business but you're actually talking to an AI-techbro whose "software-hardware hybrid product" and "incorporated company" has exactly zero revenue after it was prompted into existence in the hope that it will make some money before other people realise they could prompt the same thing for less.
Cope harder. Try harder to demoralize. It is not going to work.
I had heard HackerNews had some of this loser bunch of personalities. I didn't expect them to show themselves so soon.
But hey, I realized "AI bashing" articles are the best places to find these gate-keepers. Makes sense now.
Vibing a product into existence without needing any development knowledge or experience just means you now have a "product" that can't really be sold for money.
---
Now you want to have a "discourse"? No thanks.
Honestly, I want you to live with your assumptions and beliefs. No more goodwill from my side to your username.
Why is this? Are you not proud that you can produce products without possessing any skill?
This is my last response to you. Expended your credits.
English alphabets came into my education at the age of 10. I got my first computer at the age of 21. I began speaking broken English around the age of 23. Proper internet at the age of 25 or so.
Not to mention, my native language doesn't have programming books, even today.
Of course, an avid reader and Science nerd here. Curiosity and tinkering never stopped.
In my country, english is hardly anyones first language, but its' mandatory in schools so I've never had the experience with having to find knowledge but its gate-kept behind a translation wall.
Absolutely, I understand what you're saying.
One of the things people miss out, in most of the discussions is that they think "if you were really serious, you would have figured it out". I agree with that in most instances but language and skill acquisition is a complex process as everyone knows.
English being the de-facto reservoir of programming knowledge and applications, it takes substantial amount of time and effort to cross the threshold of understanding and transference.
In any case, I'm an eternal optimist and I believe in action. It was a great experience listening to people's opinion here and I was kind of shocked to find that some of them are so siloed in their chambers, that's interesting nonetheless.
Now AI lets you write code using libraries whose documentation you can't even read? How is this a win?
> ...
> Comes in GPT-3.5, accelerated my learning.
So now you can code? If I sat you in front of a computer with no internet and no GPU but your choice of IDE, you would actually be able to produce a product?
You're looking at a specimen of one of the most rattled species nowadays. It is so fun finding them under these articles.
Their last attempts at finding inner-validation.
What a waste.
Was that claim not true?
Honestly, I want you to live with your assumptions and beliefs. No more goodwill from my side to your username.
No discourse - you made a claim, but it appears now that your claim is untrue.
It's faster for you to answer the question than to dodge it: Have you actually learned what you claimed you did?
I don't want to shit on what you've done but you're coming in WAY too hot for how trivial your work is and how inflated your description of it is.
You're demanding humbleness while being extremely proud. Check yourself.
Anyone wondering about my proficiency, I can code without internet or AI help. But it takes enormous amount of time and mistakes.