48 ms·

Not an expert here, just speaking from experience as a working dev. I don’t think AI is going to replace my job as a software developers anytime soon, but it’s

by ednite 1y ago

Not an expert here, just speaking from experience as a working dev. I don’t think AI is going to replace my job as a software developers anytime soon, but it’s definitely changing how we work (and in many cases, already has). Personally, I use AI a lot. It’s great for boilerplate, getting unstuck, or even offering alternative solutions I wouldn’t have thought of. But where it still struggles sometimes is with the why behind the work. It doesn’t have that human curiosity, asking odd questions, pushing boundaries, or thinking creatively about tradeoffs. What really makes me pause is when it gives back code that looks right, but I find myself thinking, “Wait… why did it do this?” Especially when security is involved. Even if I prompt with security as the top priority, I still need to carefully review the output. One recent example that stuck with me: a friend of mine, an office manager with zero coding background, proudly showed off how he used AI to inject some VBA into his Excel report to do advanced filtering. My first reaction was: well, here it is, AI replacing my job. But what hit harder was my second thought: does he even know what he just copied and pasted into that sensitive report? So yeah, for me AI isn’t a replacement. It’s a power tool, and eventually, maybe a great coding partner. But you still need to know what you’re doing, or at least understand enough to check its work.

ChrisMarshallNY 1y ago
I suspect that we will be seeing some very creative supply chain compromises. If you can train AI to insert calls to your malware server, in whatever solutions it provides, that's a huge win.
- ednite 1y ago
  Totally agree! That’s where I think cybersecurity experts and system administrators deserve a lot of credit too. AI might help automate some of the work, but it’s also conjuring up threats that are way more complex and sneaky. Hopefully, countermeasures and AI-powered defense tools can keep up. It's going to be some type of an arms race, for sure.
- roywiggins 1y ago
  https://www.techradar.com/pro/security/ai-hallucinated-names-resembling-popular-libraries-created-for-slopsquatting-attacks https://www.techradar.com/pro/security/ai-hallucinated-names...
tines 1y ago
I would agree with you, but the people making the decision to fire or keep you don’t care about quality, nor do they care about understanding AI or its limitations. If AI mostly does kinda the right thing 70% of the time but saves the company $80k a year, that’s a no-brainer. We’re being Pollyanna-ish to think that anyone cares about the things we care about, that you mention in your post. If firing you saves $1.50 a year, they’ll do it.
- ednite 1y ago
  Fair enough, but what happens when those same companies start realizing it’s not just about reduced quality, but also security risks and costly errors? At some point, the savings get wiped out by the consequences. Do they go back to hiring human expertise then? I totally agree though, the business mindset of saving a buck often outweighs everything else. I’m actually going through something similar right now with a client being swayed by a so-called “AI expert” just to cut costs. But that’s a whole other story.
  - keiferski 1y ago
    They go bankrupt, get acquired, or just hope that no one notices their security mistakes. Admitting mistakes and correcting them directly is not a common thing for CEOs to do.
    - ednite 1y ago
      I guess that's what makes it scary. Good point. thanks.
    - soraminazuki 1y ago
      Just like how Equifax faced consequences for its massive data breach? Oh wait.
  - tines 1y ago
    > security risks and costly errors? I hope that you're right, but the problem is that the regulatory bodies are captured by the players that they are supposed to regulate. Can you name a time in recent history where a company had to pay a penalty for a harmful action, either intentional or neglectful, that exceeded the profit they gained from the action?
lambdasquirrel 1y ago
The AI I use at work can’t set up IAM correctly, didn’t even know it needed to, let alone associate said IAM principal with the correct k8s RBAC groups. I do appreciate that it ground through a lot of boilerplate, but I’m concerned that it’s like buying a new Sony or Leica camera as a beginner photographer. As with your non-coder VBA friend, it might make them think they’re a lot more skilled than they really are. And this is specifically why I’ve never touched VBA.
- FirmwareBurner 1y ago
  >The AI I use at work can’t set up IAM correctly Can you? :D Based on my favorite quote from the I, Robot(2004) movie with Will Smith when he got roasted by a robot: W.S.: "You are just a machine. An imitation of life. Can a robot write a symphony? Can a robot turn a canvas into a beautiful masterpiece?* Robot: "Can you?" Which I think applies to a lot of anti-AI sentiments. Sure, the AI doesn't know how to do X, Y, Z but then most people also don't know X, Y and Z. Sure, people can learn how to X, Y and Z, but then so can an AI. ChatGPT couldn't initially draw anime, now they trained it to do that and it can. Similarly it can also learn how to set up IAM correctly if they bother to train it well enough for that task. But for now, like you discovered, we're far away form an AI that's universally good at everything but I expect specialisation will arrive sooner than later.
  - bigstrat2003 1y ago
    > Can you? Yes, I can. In fact, the entire reason I think AI is not a useful tool (contrary to the hype) is that it can't do many things I find easy to do. I'm certainly not going to trust it with the things I find hard to do (and therefore can't check it effectively) in that case! For example, a month or two ago I was trying to determine if there's a way to control the order in which CloudFormation creates+deletes objects when a replacement is needed. AI (including both ChatGPT and AWS' own purpose built AI) insisted yes, hallucinating configuration options that straight up don't exist if you try to use them. The ability to produce syntactically valid configuration files (not even necessarily correctly doing what you want them to) should be table stakes here, but I find that AI routinely can't do it.
diggan 1y ago
> Even if I prompt with security as the top priority LLMs do really poorly with general statements like that, so not sure it's unexpected. If you put "Make sure to make it production worthy", you'll get as many different answers as there are programmers, because not even us human programmers agree what that really means. Same for "Make sure security is the top priority", most programmers would understand that differently. If you instead spell out exactly what behavior you expect from something like that (so "Make sure there are no XSS's", "Make sure users can't bypass authentication" and so on), you'll get a much higher probability it'll manage to follow those.
- ednite 1y ago
  Totally agree. I was simplifying for discussion’s sake, but yeah, I learned the hard way that vague prompts will lead you that rabbit hole. If you’re not crystal clear, you get everything but what you actually wanted. These days, I make sure every “t” is crossed and every “i” dotted when giving instructions. Good point, definitely a lesson worth repeating.
- kalleboo 1y ago
  Although the promise of the latest generation of "thinking" models is that they're supposed to do that themselves - in the internal chat thread in their "thought" process they're supposed to go "what does secure mean for a web app", list those items (which it can already easily do if you prompt it separately) and then go through that list
reverendsteveii 1y ago
I feel like I ran into a example use case yesterday that really illustrates how I work with AI as a dev. It's a simple db entity, with a field called "parent" that is the id of another entity of the same type. Theoretically there can be any number of parents (though practically w the data we're modeling we don't ever expect more than 3 levels). Classic case for recursion, right? So I whip up a quick method that goes something like public void getEntityWithParents(List<DBEntity> entityList, String id) { DBEntity entity = dao.getById(id); entityList.add(entity); if (entity.getParent() != null) { getEntityWithParents(entityList, entity.getId()); } } Because I'm working from a make it work first then make it efficient perspective I realize that I'm making a lot of DB calls there. So I pop open gemini, copy/paste that algorithm and ask it "How can I replace this method with a single call to the database, in postgresql?" and it gives me WITH RECURSIVE parent_chain AS (SELECT id, name, parent, 0 AS depth FROM table WHERE id = :id UNION ALL SELECT i.id, i.name, i.parent, pc.depth + 1 FROM fam.dx_concept i INNER JOIN parent_chain pc ON i.id = pc.parent) SELECT id, name, parent FROM parent_chain ORDER BY depth; That might have been a day or two of research, instead it was 5 minutes to come up with a theory and an hour or so writing tests. Gemini saved the day there. But it wasn't able to determine what needed done (minimizing DB calls), only how to do it, and it wasn't able to verify correctness. That's where we'll fit in in all of this: figuring out what to do and then making sure we actually did it.
- ednite 1y ago
  This is a perfect example of how AI can be a powerful partner in development when you already know what you’re trying to solve. Feels like the real sweet spot right now is: humans define the goal and validate the work, AI just helps fill in the middle.
  - reverendsteveii 1y ago
    if they make an AI that lets you format code readably on HackerNews we are all sauteed
    - Jensson 1y ago
TiredOfLife 1y ago
> just speaking from experience as a working dev. That's your problem. You haven't spent the past couple of years trying to get a junior job.
- ednite 1y ago
  Fair point, and I appreciate your comment. I’ve been speaking from the perspective of someone already working in the field, but let me offer another angle. Outside of coding, I’m also a new blogger and writer trying to publish articles and novels in a world already flooded with AI-generated content. In that sense, I’m in a similar position to a junior dev, just in a different domain. As I seriously consider shifting more into writing than coding, I know I’ll be facing similar competitive pressures with AI. Honestly, I have no idea where that road will lead. What I do know is that I’ll try to take advantage of every tool and opportunity available to help me succeed. For junior devs, I’d say: keep knocking on doors and focus on how AI can support your employer, not threaten either party. Funny enough, just a few weeks ago I watched an "AI expert" with no real dev experience sell a vision full of buzzwords to business leaders, and they ate it up. Hats off to him. It reminded me that today, it's not just about skills anymore. It’s about how you communicate, connect, and present your value. Sharpen your communication and people skills. Like it or not, AI is rapidly taking over the technical part. We have to stand out in the human part. Just my two cents. And for the record, please know that I don’t consider myself an expert in any of this, just speaking from experience and opinions. Hope it helps and good luck!
socalgal2 1y ago
> So yeah, for me AI isn’t a replacement. I agree with everything you wrote. The question is, what about in 6 months? 2 years? 4 years? Just a year ago (not sure the exact timeline) we didn't have the systems we have today that will edit multiple files, iterate over compilation errors and test failures, etc.... So what's it tomorrow?
- ednite 1y ago
  That’s the part that keeps me up at night at times. I’m already seeing my workflow shift fast, and the next 6–12 months feel even more unpredictable. Staying adaptable feels like the only real option, but I get that even that might not be enough for everyone.
- yifanl 1y ago
  What if in 6 months earthquakes sinks the entire lower 48?
  - ednite 1y ago
    I guess AI will be the least of our worries.
rubslopes 1y ago
> Even if I prompt with security as the top priority... A bit of a counterpoint: I've been programming for 12 years, but only recently started working in webdev. I have a general understanding of cybersecurity, but I never had to actively implement security measures in my code until now—and my boss is always pushing for speed. AI has been incredible in that regard. I can ask it to review my code for vulnerabilities, explain what they are, and outline the pros and cons of each strategy. In just a month, I've learned a lot about JWT, CSRF, DDoS protection, and more. In another world, my web apps would be far more vulnerable.
- wyclif 1y ago
  In that sense, AI is a tremendous learning tool. You could have learned about those subjects before, but it would have taken exponentially longer to search, scan, appropriate, and integrate them together to form a potential solution to a real-world problem.
geoka9 1y ago
> It's great for... getting unstuck That one I never considered until it happened to me. It's funny that the AI-provided implementation was mostly off, but it was a start. "Blank canvas paralysis" is a thing.

by ednite 1y ago

ChrisMarshallNY 1y ago

I suspect that we will be seeing some very creative supply chain compromises. If you can train AI to insert calls to your malware server, in whatever solutions it provides, that's a huge win.

ednite 1y ago
Totally agree! That’s where I think cybersecurity experts and system administrators deserve a lot of credit too. AI might help automate some of the work, but it’s also conjuring up threats that are way more complex and sneaky. Hopefully, countermeasures and AI-powered defense tools can keep up. It's going to be some type of an arms race, for sure.
roywiggins 1y ago
https://www.techradar.com/pro/security/ai-hallucinated-names-resembling-popular-libraries-created-for-slopsquatting-attacks https://www.techradar.com/pro/security/ai-hallucinated-names...

tines 1y ago

I would agree with you, but the people making the decision to fire or keep you don’t care about quality, nor do they care about understanding AI or its limitations. If AI mostly does kinda the right thing 70% of the time but saves the company $80k a year, that’s a no-brainer. We’re being Pollyanna-ish to think that anyone cares about the things we care about, that you mention in your post. If firing you saves $1.50 a year, they’ll do it.

ednite 1y ago
Fair enough, but what happens when those same companies start realizing it’s not just about reduced quality, but also security risks and costly errors? At some point, the savings get wiped out by the consequences. Do they go back to hiring human expertise then? I totally agree though, the business mindset of saving a buck often outweighs everything else. I’m actually going through something similar right now with a client being swayed by a so-called “AI expert” just to cut costs. But that’s a whole other story.
- keiferski 1y ago
  They go bankrupt, get acquired, or just hope that no one notices their security mistakes. Admitting mistakes and correcting them directly is not a common thing for CEOs to do.
  - ednite 1y ago
    I guess that's what makes it scary. Good point. thanks.
  - soraminazuki 1y ago
    Just like how Equifax faced consequences for its massive data breach? Oh wait.
- tines 1y ago
  > security risks and costly errors? I hope that you're right, but the problem is that the regulatory bodies are captured by the players that they are supposed to regulate. Can you name a time in recent history where a company had to pay a penalty for a harmful action, either intentional or neglectful, that exceeded the profit they gained from the action?

lambdasquirrel 1y ago

The AI I use at work can’t set up IAM correctly, didn’t even know it needed to, let alone associate said IAM principal with the correct k8s RBAC groups. I do appreciate that it ground through a lot of boilerplate, but I’m concerned that it’s like buying a new Sony or Leica camera as a beginner photographer. As with your non-coder VBA friend, it might make them think they’re a lot more skilled than they really are. And this is specifically why I’ve never touched VBA.

FirmwareBurner 1y ago
>The AI I use at work can’t set up IAM correctly Can you? :D Based on my favorite quote from the I, Robot(2004) movie with Will Smith when he got roasted by a robot: W.S.: "You are just a machine. An imitation of life. Can a robot write a symphony? Can a robot turn a canvas into a beautiful masterpiece?* Robot: "Can you?" Which I think applies to a lot of anti-AI sentiments. Sure, the AI doesn't know how to do X, Y, Z but then most people also don't know X, Y and Z. Sure, people can learn how to X, Y and Z, but then so can an AI. ChatGPT couldn't initially draw anime, now they trained it to do that and it can. Similarly it can also learn how to set up IAM correctly if they bother to train it well enough for that task. But for now, like you discovered, we're far away form an AI that's universally good at everything but I expect specialisation will arrive sooner than later.
- bigstrat2003 1y ago
  > Can you? Yes, I can. In fact, the entire reason I think AI is not a useful tool (contrary to the hype) is that it can't do many things I find easy to do. I'm certainly not going to trust it with the things I find hard to do (and therefore can't check it effectively) in that case! For example, a month or two ago I was trying to determine if there's a way to control the order in which CloudFormation creates+deletes objects when a replacement is needed. AI (including both ChatGPT and AWS' own purpose built AI) insisted yes, hallucinating configuration options that straight up don't exist if you try to use them. The ability to produce syntactically valid configuration files (not even necessarily correctly doing what you want them to) should be table stakes here, but I find that AI routinely can't do it.

diggan 1y ago

> Even if I prompt with security as the top priority LLMs do really poorly with general statements like that, so not sure it's unexpected. If you put "Make sure to make it production worthy", you'll get as many different answers as there are programmers, because not even us human programmers agree what that really means. Same for "Make sure security is the top priority", most programmers would understand that differently. If you instead spell out exactly what behavior you expect from something like that (so "Make sure there are no XSS's", "Make sure users can't bypass authentication" and so on), you'll get a much higher probability it'll manage to follow those.

ednite 1y ago
Totally agree. I was simplifying for discussion’s sake, but yeah, I learned the hard way that vague prompts will lead you that rabbit hole. If you’re not crystal clear, you get everything but what you actually wanted. These days, I make sure every “t” is crossed and every “i” dotted when giving instructions. Good point, definitely a lesson worth repeating.
kalleboo 1y ago
Although the promise of the latest generation of "thinking" models is that they're supposed to do that themselves - in the internal chat thread in their "thought" process they're supposed to go "what does secure mean for a web app", list those items (which it can already easily do if you prompt it separately) and then go through that list

reverendsteveii 1y ago

I feel like I ran into a example use case yesterday that really illustrates how I work with AI as a dev. It's a simple db entity, with a field called "parent" that is the id of another entity of the same type. Theoretically there can be any number of parents (though practically w the data we're modeling we don't ever expect more than 3 levels). Classic case for recursion, right? So I whip up a quick method that goes something like public void getEntityWithParents(List<DBEntity> entityList, String id) { DBEntity entity = dao.getById(id); entityList.add(entity); if (entity.getParent() != null) { getEntityWithParents(entityList, entity.getId()); } } Because I'm working from a make it work first then make it efficient perspective I realize that I'm making a lot of DB calls there. So I pop open gemini, copy/paste that algorithm and ask it "How can I replace this method with a single call to the database, in postgresql?" and it gives me WITH RECURSIVE parent_chain AS (SELECT id, name, parent, 0 AS depth FROM table WHERE id = :id UNION ALL SELECT i.id, i.name, i.parent, pc.depth + 1 FROM fam.dx_concept i INNER JOIN parent_chain pc ON i.id = pc.parent) SELECT id, name, parent FROM parent_chain ORDER BY depth; That might have been a day or two of research, instead it was 5 minutes to come up with a theory and an hour or so writing tests. Gemini saved the day there. But it wasn't able to determine what needed done (minimizing DB calls), only how to do it, and it wasn't able to verify correctness. That's where we'll fit in in all of this: figuring out what to do and then making sure we actually did it.

ednite 1y ago
This is a perfect example of how AI can be a powerful partner in development when you already know what you’re trying to solve. Feels like the real sweet spot right now is: humans define the goal and validate the work, AI just helps fill in the middle.
- reverendsteveii 1y ago
  if they make an AI that lets you format code readably on HackerNews we are all sauteed
  - Jensson 1y ago

TiredOfLife 1y ago

> just speaking from experience as a working dev. That's your problem. You haven't spent the past couple of years trying to get a junior job.

ednite 1y ago
Fair point, and I appreciate your comment. I’ve been speaking from the perspective of someone already working in the field, but let me offer another angle. Outside of coding, I’m also a new blogger and writer trying to publish articles and novels in a world already flooded with AI-generated content. In that sense, I’m in a similar position to a junior dev, just in a different domain. As I seriously consider shifting more into writing than coding, I know I’ll be facing similar competitive pressures with AI. Honestly, I have no idea where that road will lead. What I do know is that I’ll try to take advantage of every tool and opportunity available to help me succeed. For junior devs, I’d say: keep knocking on doors and focus on how AI can support your employer, not threaten either party. Funny enough, just a few weeks ago I watched an "AI expert" with no real dev experience sell a vision full of buzzwords to business leaders, and they ate it up. Hats off to him. It reminded me that today, it's not just about skills anymore. It’s about how you communicate, connect, and present your value. Sharpen your communication and people skills. Like it or not, AI is rapidly taking over the technical part. We have to stand out in the human part. Just my two cents. And for the record, please know that I don’t consider myself an expert in any of this, just speaking from experience and opinions. Hope it helps and good luck!

socalgal2 1y ago

> So yeah, for me AI isn’t a replacement. I agree with everything you wrote. The question is, what about in 6 months? 2 years? 4 years? Just a year ago (not sure the exact timeline) we didn't have the systems we have today that will edit multiple files, iterate over compilation errors and test failures, etc.... So what's it tomorrow?

ednite 1y ago
That’s the part that keeps me up at night at times. I’m already seeing my workflow shift fast, and the next 6–12 months feel even more unpredictable. Staying adaptable feels like the only real option, but I get that even that might not be enough for everyone.
yifanl 1y ago
What if in 6 months earthquakes sinks the entire lower 48?
- ednite 1y ago
  I guess AI will be the least of our worries.

rubslopes 1y ago

> Even if I prompt with security as the top priority... A bit of a counterpoint: I've been programming for 12 years, but only recently started working in webdev. I have a general understanding of cybersecurity, but I never had to actively implement security measures in my code until now—and my boss is always pushing for speed. AI has been incredible in that regard. I can ask it to review my code for vulnerabilities, explain what they are, and outline the pros and cons of each strategy. In just a month, I've learned a lot about JWT, CSRF, DDoS protection, and more. In another world, my web apps would be far more vulnerable.

wyclif 1y ago
In that sense, AI is a tremendous learning tool. You could have learned about those subjects before, but it would have taken exponentially longer to search, scan, appropriate, and integrate them together to form a potential solution to a real-world problem.

geoka9 1y ago

> It's great for... getting unstuck That one I never considered until it happened to me. It's funny that the AI-provided implementation was mostly off, but it was a start. "Blank canvas paralysis" is a thing.