At this point I literally have no idea what modern coding is supposed to look like. I still open my editor, type stuff in and run make. All these tools are released daily and they just fly over my caveman head.
I like writing code too. What I don't like doing is debugging code that just isn't doing what I want. Debugging CSS layout issues in Safari vs Chrome vs Firefox? Screw that...agent do it. Debugging layout issues in a React Native mobile app on Android on that one weird screen layout phone? Screw that....agent do it.
I don't understand what you want to hear. You can still code manually all you want, but people will not pay you for doing that in the near future. You can still do it for your personal projects.
I don't understand what you want to hear ... in the near future.
Interjecting, here. Perhaps not an academic repeat of five years ago; is this an eon for Technology or not long enough? I've lost track. Anyway, I appreciate the optimism.
An anecdote, from the ground floor: this industry has shown it's paying leagues of on-site hands to install systems, at scale, from USB keys. PXE products have existed for decades; we're writing one [to adopt] when we could take one off the shelf. These artisanal OS installations are too rich for my blood. Or, they would be, if this modernization didn't happen to be the code I'm still writing by hand. Half a decade in (or several) depending on perspective.
I'll believe things are progressing as you say... when this type of mismanagement stops happening like clockwork. Want more chatbots/datacenters? Catch up to the mid-1980s; BOOTP 'just' dropped.
Remember six months ago when linkedin people were adamant you needed to keep on top of everything not to get "left behind"? Agents, skills, mcps, swarms ...
If anything, my setup has gotten ever so simpler as models got better.
Real question here, as I'm struggling with this and I'm not finding a lot of evidence to support my inclination (that you are correct).
Why would this person be more skilled? And more importantly, what does it mean to be "skilled" in the era of AI?
I'm trying to be open-minded to both sides, but I have senior engineers on my team with a fleet of agents, shipping code they don't fully understand, and... everything is fine. Stuff still breaks, we fix it. Clients ask "How does XYZ work?" and we say, "Not sure, let us get back to you." And they're totally fine with that answer.
Everything within me screams that this is wrong and we should have intimate familiarity with the system, but I cannot find any evidence that the path of "running a team" instead of "building engineering skill" is not going to work out fine in the long-run. (Yes, my comparison between "team" and "engineering skill" is a bit rough but I'm not interested in the semantics.)
All the evidence that I'm seeing in my own business, with senior engineers meeting clients' needs with autonomous agentic teams is that it's all fine, and while there are gaps, it's actually okay. Clients are happy. Software works 80-90% of the team exactly as it should. Bugs are fixed quickly. We ship more in 3 months than we previously could've shipped in a year. I'm baffled and frankly not as advanced as my devs (I still ask questions one by one to my agent, like a Real Caveman!)
It makes me wildly uncomfortable so I'm trying to wrestle with this. I want your statement to be true, but I just don't see evidence of that.
Software works 80-90% of the team exactly as it should.
Either you work in a non-standard field, or your standards for software are way too low.
What software gets to fail 80% of the time? If your ecommerce checkout software fails 1 in 5 times, you don't have an ecommerce business at all, you are dead in the water. If your airline booking software fails for 1 in 5 customers, you are out of business.
99.99% reliability still means 1 in 10000 customers are experiencing an error, which at scale can mean thousands to millions per day. That's completely unacceptable for any software I've ever been involved with writing.
I think you might be taking that number a bit too literally. What I mean is, it works enough that no one is complaining. But, all software has bugs. Some are logic/syntax bugs, and some are intent. "It should do this, but actually does that."
The sum of ALL of these classes of bugs is such that our clients haven't complained about the software being particularly faulty. It's impossible to tell if it's more or less buggy than it would've been written entirely by hand, but I suspect it's less buggy than it would be if written by hand.
I'm with you on this. It feels like it should matter, but everything is pointing to "it doesn't matter as much as you guys say that it matters". I had the same experience at work. I thought we would have way more bugs and problems, yet all I see is the same as ever, but faster. It's eye-opening, honestly
We are seeing the same. We are moving significantly faster and ship more complex stuff, and there are still bugs as they were in the before times, but nothing catastrophic, no gremlins in the machine. Our main worry is about process and communication, not the actual coding really.
I think the biggest issue with LLM generated code is that it's happy to just keep plugging away making a bigger and bigger rat's nest.
I spend a lot of time reviewing code and the LLM code works for what the implementer wanted and from the outside things function the way people expected, but inside the code is becoming more and more knotted and incomprehensible.
Now maybe you could point claude at it and say "make the code easy for people to understand too" and no one is bothering to do that. Maybe it's a skill issue and the people around me should be prompting better or revising better.
The only thing I know is that I'm seeing it happen. Code quality is going down, comprehension is not just missing because people didn't author the code, it's getting worse because even when they go to read the code it's become too complex and weird. So now they have to rely on LLMs to tell them what the code they are responsible for is doing and how.
The thing that remains to be seen is if comprehensibility will matter. I can't look at the assembly the compiler outputs and comprehend it and it doesn't bother me. LLM advocates argue the same will become true for code, it won't matter if you understand the code because it only matters that the LLM can comprehend it and keep making it do the things you want.
I'm less inclined to believe that being able to reason about your codebase and make changes to it is something I really want to put into a bunch of hyperscaler company hands. Enshittification tells me that this honey moon of affordable intelligence will reach an end at some point and a lot of companies are going to clamp down on more expensive tokens and a lot of engineers are going to find that they have both let their code comprehension skills atrophy and aren't permitted to take every little question and change to the LLM because it's too costly.
They will crack open the codebase, every changeset with a full comprehensive page of LLM-speak explaining how it addresses the problem, every changeset in isolation reasonable enough, and an end state that is something only an LLM can love.
I dunno dude, it's getting scary fast. You tried a fable/5.6sol maximum effort on a chunky task yet?
Maybe in a few years being a good coder will be like being a good taxi driver with self driving. You know the best spots and the ai can't drive the most unusual roads but effectively 95% of all driving is self driving.
Interesting that you use self driving as the metaphor here, when we've been told for years that self driving will be taking over the roads any day now... and yet...
I use waymo all the time? It's literally here. I massively prefer it to a cab. Only question left is how long it takes to scale and make cheap.
Prolly what, 30yrs? Places like SV or other advanced locations will hit majorty faster than that. The rest of the world ~50yrs.
Hype is fast but tech is a decades rollout. AI will hit its stride in 5-10yrs. Which is scary because it's already stupidly powerful. Saturation of AI though is still a 20yr+ horizon.
Letting it rip and give it full infra, db, dns, all the keys to the castle, drinking from a telemetry firehose the whole time, a huge tangled web of integrations that all kind of works. I think just offloading as much as possible is the state of things.
I open my editor, I need to implement a feature, I tell chatGPT what I need, it reviews my code and provides suggestions. I then tell it to implement it. I then tell it to write tests. I review the code.
I test it. I have another AI do code review. I push it to test, then after tests to production.
For some common tasks, that I don't find interesting it is great. I do a lot of CRUD, I honestly cannot imagine writing these apps without AI.
It made curious and freed my brain to think about other things. I am now playing with hardware and plan on launching a niche product. This would not be possible if I 100% committed to programming without AI.
Comments
At this point I literally have no idea what modern coding is supposed to look like. I still open my editor, type stuff in and run make. All these tools are released daily and they just fly over my caveman head.
I used to be that person, and suddenly I haven’t opened an editor or written a line of code by hand in 6+ months.
Same here.
All I do now is spend all day on calls with people whose shit exploded and I'm trying to scrub it off the walls, ceiling and the customers.
Ah, so then you became a plumber!
Sewage worker
All money is green. Ensure you are being compensated sufficiently for your time.
Most of the world’s money is not, in fact, green.
Yes. No money no work.
But why? I like writing code.
Why in the world would I want to become someone that doesn't write code?
I like writing code too. What I don't like doing is debugging code that just isn't doing what I want. Debugging CSS layout issues in Safari vs Chrome vs Firefox? Screw that...agent do it. Debugging layout issues in a React Native mobile app on Android on that one weird screen layout phone? Screw that....agent do it.
I thought I loved writing code. Turns out I like building shit. Why take a week to build something I can build in a day?
I still like reviewing code, but I don’t need to physically write it
The person paying you doesn’t care about your enjoyment.
I do care about my enjoyment. Or should my existence, wants and dreams solely be defined by my current employer?
Are yours?
I don't understand what you want to hear. You can still code manually all you want, but people will not pay you for doing that in the near future. You can still do it for your personal projects.
Interjecting, here. Perhaps not an academic repeat of five years ago; is this an eon for Technology or not long enough? I've lost track. Anyway, I appreciate the optimism.
An anecdote, from the ground floor: this industry has shown it's paying leagues of on-site hands to install systems, at scale, from USB keys. PXE products have existed for decades; we're writing one [to adopt] when we could take one off the shelf. These artisanal OS installations are too rich for my blood. Or, they would be, if this modernization didn't happen to be the code I'm still writing by hand. Half a decade in (or several) depending on perspective.
I'll believe things are progressing as you say... when this type of mismanagement stops happening like clockwork. Want more chatbots/datacenters? Catch up to the mid-1980s; BOOTP 'just' dropped.
You're right, they won't be paying me to write code in the near future... They will be paying me to fix code.
Good times ahead!
Yes they will.
I feel like theres so much churn and change there's no point in trying to keep up.
Remember six months ago when linkedin people were adamant you needed to keep on top of everything not to get "left behind"? Agents, skills, mcps, swarms ...
If anything, my setup has gotten ever so simpler as models got better.
Why go on Linkedin? The "don't get left behind" people are all in this comment section.
Can’t get left behind by people who can’t write a crud endpoint or an email by themselves.
That’s fine, and you’re doing fine. In fact, in a few years you will probably be far more skilled than most “software engineers”.
Real question here, as I'm struggling with this and I'm not finding a lot of evidence to support my inclination (that you are correct).
Why would this person be more skilled? And more importantly, what does it mean to be "skilled" in the era of AI?
I'm trying to be open-minded to both sides, but I have senior engineers on my team with a fleet of agents, shipping code they don't fully understand, and... everything is fine. Stuff still breaks, we fix it. Clients ask "How does XYZ work?" and we say, "Not sure, let us get back to you." And they're totally fine with that answer.
Everything within me screams that this is wrong and we should have intimate familiarity with the system, but I cannot find any evidence that the path of "running a team" instead of "building engineering skill" is not going to work out fine in the long-run. (Yes, my comparison between "team" and "engineering skill" is a bit rough but I'm not interested in the semantics.)
All the evidence that I'm seeing in my own business, with senior engineers meeting clients' needs with autonomous agentic teams is that it's all fine, and while there are gaps, it's actually okay. Clients are happy. Software works 80-90% of the team exactly as it should. Bugs are fixed quickly. We ship more in 3 months than we previously could've shipped in a year. I'm baffled and frankly not as advanced as my devs (I still ask questions one by one to my agent, like a Real Caveman!)
It makes me wildly uncomfortable so I'm trying to wrestle with this. I want your statement to be true, but I just don't see evidence of that.
Either you work in a non-standard field, or your standards for software are way too low.
What software gets to fail 80% of the time? If your ecommerce checkout software fails 1 in 5 times, you don't have an ecommerce business at all, you are dead in the water. If your airline booking software fails for 1 in 5 customers, you are out of business.
99.99% reliability still means 1 in 10000 customers are experiencing an error, which at scale can mean thousands to millions per day. That's completely unacceptable for any software I've ever been involved with writing.
I think you might be taking that number a bit too literally. What I mean is, it works enough that no one is complaining. But, all software has bugs. Some are logic/syntax bugs, and some are intent. "It should do this, but actually does that."
The sum of ALL of these classes of bugs is such that our clients haven't complained about the software being particularly faulty. It's impossible to tell if it's more or less buggy than it would've been written entirely by hand, but I suspect it's less buggy than it would be if written by hand.
I'm with you on this. It feels like it should matter, but everything is pointing to "it doesn't matter as much as you guys say that it matters". I had the same experience at work. I thought we would have way more bugs and problems, yet all I see is the same as ever, but faster. It's eye-opening, honestly
We are seeing the same. We are moving significantly faster and ship more complex stuff, and there are still bugs as they were in the before times, but nothing catastrophic, no gremlins in the machine. Our main worry is about process and communication, not the actual coding really.
I think the biggest issue with LLM generated code is that it's happy to just keep plugging away making a bigger and bigger rat's nest.
I spend a lot of time reviewing code and the LLM code works for what the implementer wanted and from the outside things function the way people expected, but inside the code is becoming more and more knotted and incomprehensible.
Now maybe you could point claude at it and say "make the code easy for people to understand too" and no one is bothering to do that. Maybe it's a skill issue and the people around me should be prompting better or revising better.
The only thing I know is that I'm seeing it happen. Code quality is going down, comprehension is not just missing because people didn't author the code, it's getting worse because even when they go to read the code it's become too complex and weird. So now they have to rely on LLMs to tell them what the code they are responsible for is doing and how.
The thing that remains to be seen is if comprehensibility will matter. I can't look at the assembly the compiler outputs and comprehend it and it doesn't bother me. LLM advocates argue the same will become true for code, it won't matter if you understand the code because it only matters that the LLM can comprehend it and keep making it do the things you want.
I'm less inclined to believe that being able to reason about your codebase and make changes to it is something I really want to put into a bunch of hyperscaler company hands. Enshittification tells me that this honey moon of affordable intelligence will reach an end at some point and a lot of companies are going to clamp down on more expensive tokens and a lot of engineers are going to find that they have both let their code comprehension skills atrophy and aren't permitted to take every little question and change to the LLM because it's too costly.
They will crack open the codebase, every changeset with a full comprehensive page of LLM-speak explaining how it addresses the problem, every changeset in isolation reasonable enough, and an end state that is something only an LLM can love.
Or unemployed, the jury on that is still out.
Or both?
I dunno dude, it's getting scary fast. You tried a fable/5.6sol maximum effort on a chunky task yet?
Maybe in a few years being a good coder will be like being a good taxi driver with self driving. You know the best spots and the ai can't drive the most unusual roads but effectively 95% of all driving is self driving.
Still waiting to see the scary products of AI. I see cool tech demos, sure. But I'm not scared just yet.
A very large percentage of systems you use every day being built with AI?
Coding has changed butbuissness has not.
Interesting that you use self driving as the metaphor here, when we've been told for years that self driving will be taking over the roads any day now... and yet...
I use waymo all the time? It's literally here. I massively prefer it to a cab. Only question left is how long it takes to scale and make cheap.
Prolly what, 30yrs? Places like SV or other advanced locations will hit majorty faster than that. The rest of the world ~50yrs.
Hype is fast but tech is a decades rollout. AI will hit its stride in 5-10yrs. Which is scary because it's already stupidly powerful. Saturation of AI though is still a 20yr+ horizon.
1.5 million people use it for 90%+ of their drives
There are >23 million millionaires in the US and most people avoid generalizations based on them.
Letting it rip and give it full infra, db, dns, all the keys to the castle, drinking from a telemetry firehose the whole time, a huge tangled web of integrations that all kind of works. I think just offloading as much as possible is the state of things.
I open my editor, I need to implement a feature, I tell chatGPT what I need, it reviews my code and provides suggestions. I then tell it to implement it. I then tell it to write tests. I review the code.
I test it. I have another AI do code review. I push it to test, then after tests to production.
For some common tasks, that I don't find interesting it is great. I do a lot of CRUD, I honestly cannot imagine writing these apps without AI.
It made curious and freed my brain to think about other things. I am now playing with hardware and plan on launching a niche product. This would not be possible if I 100% committed to programming without AI.