> I simply know that I can solve any engineering problem (that I can come up with) with the help of artificial intelligence. I've tried it a few times — it works.
I think you just need to push yourself to do harder problems. For me, I pushed pretty aggressively into the hardware space, building things as a hobby that I originally felt would be far too time-consuming. This is an area where AI can certainly help accelerate but asking it to make small changes can be an expensive (token wise) proposition, so it forces you think through all the critical decisions and research up-front, and let AI do the boring scut-work.
There's still tons of pure software engineering problems where iterations are time consuming, or there's no obvious solution, like physics simulations, DSP, computer graphics... I can trivially up the difficulty on any tasks that it can easily solve.
LLMs enable me to iterate quicker, accelerate and parallelise, research, scour the literature and integrate cross-disciplinary knowledge, easily implement more complicated algorithms, and are generally better at avoiding trivial mistakes.
Before LLMs, I can't believe how many hours I wasted debugging computer graphics or computer vision algorithms only to realise the problem was my handling of coordinates, flipping x and y, or one program needing the y axis flipped, or typos when copying the algorithm from equation to code. I have no nostalgia for that era.
Maybe you're not actually solving a problem, but revisiting a solved problem? Maybe you need to pursue bigger problems, or stop playing this version of the game? How can the title be "Challenge is Gone", then the post be about how what you used to do is (you believe) no longer relevant, and you're not sure what's next - isn't this the answer you're looking for?
Yes. Hardware is hard. If you even design… what, a simple led flasher, you will go through several iterations in production based on your purpose (every target audience has an almost endless list of constraints that only become clear after making and failing).
On of my planned retirement projects has been to make a new TRS-80 Model 100 slab computer and that includes all the software (ie no Linux). It feels like some of the models are good enough that I’m a few hundred dollars in tokens away from a prototype (I’d make a simulator first).
I mean its really small things, like the screws I bought are slightly out of spec and dont fit perfectly and loosen during shipping. Sure AI can vet for recommended parts and vendors, but what if these are not available ‘in your area’?
Yes but doing it like that helps the AI labs produce better AI. It's going fast now. Over the past few years AI has learned _faster_ to become a good software engineer than CS/SE college students do over a standard 4 year curriculum.
But these AI labs are not Linus. So, they'll just keep on gobbling up, but never give. And, all the new comers, are going to be a loser, to feed some billionaire.
I am roughly the same age and I grapple with many of the same ideas, although I'm still pretty depressed at this point. I've been coding for 25 years, I'm quite good at it, it's a facet of my personality. Seeing this skillset commoditized within a year or so has been a brutal experience. Payback for all the jobs we replaced with software, I suppose.
I do feel like I should be thankful to be alive right now though. I witnessed the evolution from Kirby on the Game Boy to Half Life 2 to VR to LLMs. The tools we use today were science fiction only a few years ago.
I just wish society was in a better state to make this tech a greater good instead of a tool for advertisement, warfare and wealth concentration. I feel a deep sense of loss because of this. 20 years ago I had no doubt that technological progress was an obviously good thing, that it would improve the lives of everyone. Do I still believe this? I'd like to, at least.
I feel that as a field, we overfitted towards economic value maximization (which, to be fair, is the obvious thing to do) while leaving important problems behind.
Accessibility, sanitary infrastructure (in developing countries), generally improving health (including making it cheaper), and environment/wildlife preservation being some good examples of largely unsolved problems that technology still holds the key to dramatic improvement.
This might be the right time to start working on these things. Ironically, warfare, advertisement, and wealth concentration helped advance technology rapidly, making it easier to tackle other problems.
I appreciate the comment because it resonates so closely to how I feel. Honestly, I could have written nearly the same thing. Thanks for sharing. I still do hang on the idea that there is still good we can bring into the world.
You (not responding to you directly simias) are truly delusional if you think what's happening isn't extremely concerning and dystopian in a lot of ways.
And there will be a lot of ego death, esp. with people deep in a lot of kowledge work.
But if you claim that AI is absolutely a waste and bad for humanity I feel sorry for you if you can't see how incredible it is that we have this new technology. -That it will most likely solve some very core issues humans are working on, and that it is a kind of evolutionary moment in human history, creating artificial intelligence. I feel sorry that you can't see past your own nose to see the bigger picture in that sense.
I tend to think big changes will always make things better and worse at the same time. If you look at the long horizon of history (100s of years) it appears that it makes things better on balance. But who really knows for sure what will happen.
I think at this point I am mostly in AI camp. The recent advances in reverse engineering automation ( and subsequent panic -- and upcoming banhammer - coming from owners and shareholders of previously available money makers ) made me recognize it is a real shift and not just an incremental improvement.
People are scared. I suppose they should be. A lot of people have been invested in ensuring that things stay roughly the way they are. I am not sure that was ever sustainable, but now it seems like an impossible dream.
Cat is out of the bag. Milk is spilt. Hanninal crossed the alps.
Am I worried? Yes. But it is still fucking amazing to watch it unfold.
Like problem solving, art, music, writing, and other intellectual challenges. And doing it well enough that you make an impact on the people around you.
Programming is one such creative endeavor, but they're all under attack.
Personally, when it comes to media and creative things, I want to hear from other humans. It could be other humans using AI as a tool, but I'm not interested in the Meta AI account slop they were promoting. Or the Spotify algorithm optimized playlist artists.
I don't think everyone will agree with me, and we might have truly AI generated movie and pop stars, but I think most people won't want synthetic art- it's missing the point of what art is for, for expressing something about humans.
Right now it's easy to say, because the AI generated stuff is so obviously bad. What if it were better than human output? What if the only way to tell was that the human sounded worse?
You're betting on AI continuing to be bad, or on humans rejecting it wholesale. While I wish humans would reject it, I'm not so optimistic.
I was genuinely considering whether or not I still liked programming - the excitement and risk of technical failure had dropped off. Most failures were for other organizational and process and management reasons. I respect the value of executing those aspects well, but I have no interest in becoming fluent in solving that class of problems. It just bores me. I want to build big beautiful, complex, intricate machines.
The agentic tools are, for me, a step up into tackling broader projects I had put on the shelf as "not doable without investment". I can self-fund my ideas now. Things I anticipate would have taken about a decade or so to play around with and build out as a sole developer, I'm taking on as a sole developer.
But that said, I do appreciate that we can't be sure of what exactly this churn will produce. The other factors that go along with the technical achievement aren't too rosy.
<< I can self-fund my ideas now. Things I anticipate would have taken about a decade or so to play around with and build out as a sole developer, I'm taking on as a sole developer.
This. All of a sudden, I have real personal goal timeline of things I want to build. The odd thing is, there are still hard things on that horizon, where I can see clearly things will require a team of humans, who understand how things really work under all these abstractions.
Hmm, sometimes I feel like living in a different universe. Since LLMs can solve any technical challenge (like most of you claim here) and help folks explore new fields and revive all sorts of projects, wouldn't be all seeing effect of that everywhere? Wouldn't the bar for software quality go up (and go up significantly)? Wouldn't we see new innovative apps that people are delighted to use and more than happy to pay money for? Yet when I recently challenged some friends of mine to name just ONE major app / service that became materially better (I'd argue even better at all) or a new/innovative widely adopted by masses outside of pure AI world (Claudes and GPTs), they couldn't name any. Meanwhile, I am looking around and see that banking, insurance, search, social media, cloud computing are getting more bloated, unstable, mediocre and in general of lower quality. Where is all the cool software everybody is talking about? Where are the features from your backlogs that you were finally able to burn through? I am just not seeing it in the real world. Something doesn't add up here.
In the past we had communities. They provided us with a context of meaning for our lives. We have lost that. So the context people have for their lives is either money, or just themselves.
Many people have not even experienced community these days. Imagine where you feel part of something bigger than yourself. A gestalt, where the whole is greater than the sum.
We People grew being in that situation - we would hunt, farm, fix things, make things - better not just for ourselves but for our community. It became our nature, just as it has became the nature of sheep dogs to herd.
Imagine a sheep dog who lives a life with no sheep. They have all that built into their nature and never get to be that. That is you and so many other people. So there is your challenge.
It is interesting that you say that, because it may explain my relative ease with this particular transition. Community ( as understood these days -- not exactly a neighborhood, but not just your household either ) was not really my guiding light before so its potential loss is not as pronounced to me. In other words, loss of that context does not visibly affect me.
I categorize spelling and grammar corrections as basic fact checking. I like to write how I talk and that means I sometimes leave the bad grammar because it seems more like my voice (whatever that is).
I think that’s the whole point the article/comments are making.
Many things we used to learn are becoming obsolete by AI.
A analogy (if not the best one)3 People used to be much less sedentary, and in some ways healthier, before automobiles (or working from home).
So setting aside intentional exercise became much more necessary to be less sedentary and healthier.
But it requires discipline and motivation, and many people don’t exercise enough as a result.
Before AI, learning was a necessity to get things done. We had to check our own spelling/grammar before these tools existed. Etc.
Now, learning and using these skills will be simply for the “exercise” of using our brains in ways that were necessary before AI.
Some people will continue to code ok the side “for fun” or “for exercise”, but it will no longer be a means to simply getting something done — that can be done with AI.
Any you certainly won’t be doing for a company whose primary focus is to make money.
This might be the end of personal challenges but we have so many really hard challenges that could do well with a lot of people coming together. How do we solve for poverty and the lack of clean air, water, and food in urban India (could be anywhere, but this I see daily). What could we do to get people to collaborate at scale on problems that are not currently profitable for an org? Cheap air conditioning. Senior citizen loneliness. So many problems to take up. We probably can't even think in that direction because these seem so huge now. But that's hopefully what intelligence on tap is now bringing into the tractable domain
I've tried a few things that I know are possible (multiplayer FPS with client-side prediction), but the overall challenge still persists. There is no easy way to rapidly iterate this architecture with an agent unless you already have massively deep pools of experience to draw from. Launching this system into the real world is even more challenging than building it.
For many problems, the LLMs look like they might be making progress, but once you zoom out far enough you realize it's just pushing the toothpaste around in the same tube with the cap still on. The mass of the overall system is not changing despite the appearance of localized progress.
If your primary source of identity was gatekeeping the Pantone colors of the various office furniture, then yes perhaps the challenge is now gone. But, if you were focused on mission and 'big picture' (what furniture and why), it's plausible that the challenge could actually increase in its intensity.
I don’t know man, a lot of shit I try still doesn’t work. Maybe the challenge has relocated from doing the right thing to asking the right questions and knowing what the right answer looks like — but it’s definitely not “gone”.
It works well enough that the work remaining is the sloggy manual testing and bug reporting. You're right that it doesn't work well, but the interesting part has been automated.
This is true in general: janitorial work is likely to outlive white collar work in this hell we're building.
It takes only a few seconds to discover dozens of unsolved engineering challenges in multiple fields. The only things that are going are the trivial ones. Ignoring that, LLMs bring a whole slate of their own challenges.
Challenge is not gone. Use LLMs to clear up your backlog of things you wanted to do before but couldn't, then choose something that's challenging even with the help of an LLM. Dive into a skill you thought you didn't have time to learn. Make a something you previously thought was too big to make by yourself.
I’ve been trying to get better at analyzing performance traces (like Bruce Dawson has written about on his blog), crash dumps, and assembly language in general.
AIs are still pretty bad at all this stuff.
It’s made we wonder if future AIs will get much better at machine language and will skip writing human readable source code and just dump out CPU opcodes (ie write the binary directly).
This. Running a load of things on a server at home was on my backlog for ages. I started manually then got Claude to reverse engineer the system and create a load of Ansible that we iterated on. It still needs its hand holding , and likes to pretend it doesn't know what good Ansible looks like but we got there
It also wrote a nifty remote LUKS unlock utility. So handy
I think the author (and many people, including myself) is missing something crucial.
Yes AI can solve very difficult problems without requiring you to understand enough about it.
This gives the impression that you can solve anything, create anything, and put it out there. And because this seems to be true, you feel less motivation to go and do it, because it’s so easily achievable. It seems AI is robbing your ability to differentiate yourself and push hard on a direction and build something that matters.
However, I have a feeling that is increasing over time. I believe that this is not true at all.
Even with AI, an enormous amount of energy needs to come from within one self, to truly create something meaningful that is used by other people. Just because AI can spit out code that solves a set of problems, it doesn’t mean it’s ready to be used and that it will turn out to be something people actually care and use.
If your only ambition was solve a hard problem, and the problem was purely technical, I agree you are in a bad spot, but then again this is not gone. People play single player videogames fully knowing they can finish the game and they can even hack if they wanted but they don’t, they still enjoy playing, it just means this is now perhaps a hobby rather than something you can profit from.
Creating a product however, remains difficult I believe. A product that people actually enjoy using and solves something. If anything the bar is higher, and it is very hard to differentiate. I think this is more challenging than before, not less.
So if you shift your ambition to be, I want to create something people actually enjoy and it’s useful, and I can survive the next few years and have a competitive advantage, I think this is incredibly difficult.
I do miss programming instead of reading code. But I enjoy designing and picking languages and tools. For me AI has been great (so far): I revived old projects and came up with new ones, including branching to hardware products. This would have never been achievable without AI. Not because a lack of skill. I'm sure I could have picked any of the projects and see it through. But I wouldn't have, because the time and effort wouldn't have justified the result. And I would never have been able to do so many of them. I would have needed to pick one, pour many, many hours of work into it and all the other things would need to wait. So I could have done a fraction, but in reality I would have hardly done any. So for me it's enabling.
Currently, for me the challenge has gone from can I build it, to can I build this and everything else I need, and structure it to be maintainable for the long term, so I can keep stacking bricks. This has found me leaning harder into IaC, use of and contribution to open source (for secure ownership, longevity of system and cost reduction) and structuring the development tooling around agents. That will be an ongoing process, but I feel that I'm at the point where I can move onto building what matters. I've already built a few of the tools I've dreamed of having for years (there's a couple of rough edges but it beats the tools I had before), and I think, like the author this will free me up to stay on the next step (building what matters) for good.
The main one I see is the question of how we can use an LLM as humans to produce something harder than the LLM can by itself.
I like to think of it with an analogy. Suppose I was granted the privilege of having a world renowned human expert as my personal assistant, but she had no initiative of her own. How could I collaborate with her?
I would definitely need to keep learning more to understand what she says. I would also need tools to augment my ability to follow along and she would need tools to bring knowledge down to my level of understanding.
One might argue humans are unnecessary in the above process, but I think that's just failing the challenge of keeping up. So I try
I don't know, the set of problems that remain very hard but worthwhile to solve still massively dwarfs the set of problems that has become easy to solve with the help of AI, even in a restricted domain like software engineering.
It is a failure of imagination to think that there is no longer much challenge. But I guess all humans have a problem with tunnel vision, to various extents.
Come join those of us that enjoy solving problems at work but aren’t driven by it. The water is fine, the weather is usually decent, and the existential terror of “failing” goes away eventually. We have the same problems and worries but sleep a bit better when things go wrong.
All challenge is gone? No way. Your next challenge is to find a new one.
LLMs make what was formerly hard trivial, that is true. But they cannot do everything with simple prompts, you must increase your ambition to what was formerly impossible. Try building a bigger game, more complicated software or hardware…what have you always wanted but written off as something only a multi-millionaire with a large team of domain experts could accomplish? Because you have one now.
My initial reaction as well. If you believe AI has made what you do irrelevant, change what you do. If you really want to do something different, stop playing the knowledge arms race game with a computer program. There has been "analog" LLM-like ventures (aka big, rich tech corporations) for a long time, yet they didn't previously solve all engineering problems; you can now use technology as a great equalizer, or change your focus to something AI can't even do if you don't want to play the game.
I was glad to see the direction this essay took. I agree with you: it's the old challenges that are gone and there's a whole new class of challenges (the former impossibilities) in front of us.
That's a feeling I've been also feeling for the last year or so. Turns out the means was what I cared about, and I didn't care about the ends that much. LLMs robbed me of that in a way I couldn't see coming.
I think that we as a society fail to see that the speed we built things was rarely the problem. The real shortage was in the ideas - having them, finding the motivation to act on them, sharing them, and the biggest driver was the desire of showing off, being told you are great and your idea was great.
I am afraid that by removing the ick to fix things, to challenge everything and being rewarded for that, we will end up with less innovation, not more. Why bother? Why do anything if everything is cheap, why start a new project only to be shrugged away as being yet another thing any machine could have done? Why show off when the merits are the machine's?
HN went from being a constant stream of Show HN to AI news after AI news. The most interesting projects AI have enabled are to run old stuff onto other things.
Rehashes of old ideas keep sprouting out of everywhere, keeping new ones at bay. I think we are probably entering a long winter and all that will be left is permutations of what we did before '23
Ai helps me build better, more deterministic outcomes that it itself can't. I mean I'm building a mailing list manager that blends names of couples. Yes, AI can do this but it's inconsistent. I can direct AI to code precise outcomes step-by-step. Coding is solved? Not sure, but my directing code is solved.
> I was only ever interested in engineering challenges.
When you pigeonhole yourself into only finding challenge in one field, you're bound to be disappointed eventually. While I do miss some of the challenge, these tools have given me more freedom to tackle difficult problems outside of software, both personally and professionally.
Yesterday I turned 40 and reflected on the same thing.
For most of my life the “main quest” was always obvious. I went to school, found my soul mate, got married, bought a house, had kids, paid off the mortgage, got yelled at by Howie Mandel, won the awards, achieved my career goals.
But now there’s no main quest. It’s just a whole wonderful, brilliantly rendered world of side quests for me to pick from. I’m spoiled with choice…
The thing is… main quests drive you, but you have to drive the side quests; they’re perfectly content to not get done. It’s an adjustment, and I’m optimistic that I’ll overcome it. But I’m currently trapped in the doldrums of figuring out what to do between now and when the credits roll.
Indeed; but for the author, the challenge was not to produce a good translation of anything, or to develop a good automated process for doing so. It was to bash together a website that churns out translations, without even a specified level of quality. For the author, these are all challenges of making a process to do something; not to produce any actual artifacts of quality.
Here's a challenge: see yourself from the outside.
Realize that however effective you are as a programmer now, you were once a hopeless beginner.
Then you must acknowledge that you are still a hopeless beginner, relative to the you that doesn't exist yet.
In the spirit of fellowship, I tell you:
git gud
---
And as you continue to learn, you'll see this for yourself, but I guarantee to you now that AI opened paths to writing novel software which no one has noticed yet. Software that was possible before but perhaps annoying to build, etc. Software which perhaps will only ever be imagined by someone who learned exactly the things you have, in exactly the order you did.
Never make the mistake that AI knows how to do everything just because it knows how to do everything we've done so far.
And don't think of your work as "mining" engineering problems! Forget engineering problems. Find a human problem and think about it for an entire month.
understanding of the unconscious would be entirely novel , yet such knowledge is overtly claimed here. if the author is human he may be showing signs of ai psycosis.
> I simply know that I can solve any engineering problem (that I can come up with) with the help of artificial intelligence.
I'm sorry, but I find this a bit silly. Either:
* your imagination is extremely limited (which is OK, no worries!), and the statement is true because of that, or
* your imagination is fine and the statement is just false, or
* your imagination is fine and the statement is true, in which case: what's stopping you? We all agree that there are engineering challenges, the solution to which would make the world a better place (or make you fabulously wealthy, whatever floats your boat), so get cracking if it's so easy now!
> I simply know that I can solve any engineering problem (that I can come up with) with the help of artificial intelligence. I've tried it a few times — it works.
I think you just need to push yourself to do harder problems. For me, I pushed pretty aggressively into the hardware space, building things as a hobby that I originally felt would be far too time-consuming. This is an area where AI can certainly help accelerate but asking it to make small changes can be an expensive (token wise) proposition, so it forces you think through all the critical decisions and research up-front, and let AI do the boring scut-work.
There's still tons of pure software engineering problems where iterations are time consuming, or there's no obvious solution, like physics simulations, DSP, computer graphics... I can trivially up the difficulty on any tasks that it can easily solve.
LLMs enable me to iterate quicker, accelerate and parallelise, research, scour the literature and integrate cross-disciplinary knowledge, easily implement more complicated algorithms, and are generally better at avoiding trivial mistakes.
Before LLMs, I can't believe how many hours I wasted debugging computer graphics or computer vision algorithms only to realise the problem was my handling of coordinates, flipping x and y, or one program needing the y axis flipped, or typos when copying the algorithm from equation to code. I have no nostalgia for that era.
Maybe you're not actually solving a problem, but revisiting a solved problem? Maybe you need to pursue bigger problems, or stop playing this version of the game? How can the title be "Challenge is Gone", then the post be about how what you used to do is (you believe) no longer relevant, and you're not sure what's next - isn't this the answer you're looking for?
Yes. Hardware is hard. If you even design… what, a simple led flasher, you will go through several iterations in production based on your purpose (every target audience has an almost endless list of constraints that only become clear after making and failing).
On of my planned retirement projects has been to make a new TRS-80 Model 100 slab computer and that includes all the software (ie no Linux). It feels like some of the models are good enough that I’m a few hundred dollars in tokens away from a prototype (I’d make a simulator first).
I mean its really small things, like the screws I bought are slightly out of spec and dont fit perfectly and loosen during shipping. Sure AI can vet for recommended parts and vendors, but what if these are not available ‘in your area’?
Basically: run as fast as you can to escape AI. It's coming for the challenges you enjoy.
I’m still trying to learn guitar even though I can stream all the guitar music I could ever listen to for almost nothing.
[dead]
Yeah but it's running too, and is accelerating.
Yes but doing it like that helps the AI labs produce better AI. It's going fast now. Over the past few years AI has learned _faster_ to become a good software engineer than CS/SE college students do over a standard 4 year curriculum.
But these AI labs are not Linus. So, they'll just keep on gobbling up, but never give. And, all the new comers, are going to be a loser, to feed some billionaire.
I am roughly the same age and I grapple with many of the same ideas, although I'm still pretty depressed at this point. I've been coding for 25 years, I'm quite good at it, it's a facet of my personality. Seeing this skillset commoditized within a year or so has been a brutal experience. Payback for all the jobs we replaced with software, I suppose.
I do feel like I should be thankful to be alive right now though. I witnessed the evolution from Kirby on the Game Boy to Half Life 2 to VR to LLMs. The tools we use today were science fiction only a few years ago.
I just wish society was in a better state to make this tech a greater good instead of a tool for advertisement, warfare and wealth concentration. I feel a deep sense of loss because of this. 20 years ago I had no doubt that technological progress was an obviously good thing, that it would improve the lives of everyone. Do I still believe this? I'd like to, at least.
I feel that as a field, we overfitted towards economic value maximization (which, to be fair, is the obvious thing to do) while leaving important problems behind.
Accessibility, sanitary infrastructure (in developing countries), generally improving health (including making it cheaper), and environment/wildlife preservation being some good examples of largely unsolved problems that technology still holds the key to dramatic improvement.
This might be the right time to start working on these things. Ironically, warfare, advertisement, and wealth concentration helped advance technology rapidly, making it easier to tackle other problems.
Yeah, I am also grateful to have been around to watch the twilight of humanity as we outsource our abilities.
Tech has certainly stopped being good for people.
I appreciate the comment because it resonates so closely to how I feel. Honestly, I could have written nearly the same thing. Thanks for sharing. I still do hang on the idea that there is still good we can bring into the world.
You (not responding to you directly simias) are truly delusional if you think what's happening isn't extremely concerning and dystopian in a lot of ways.
And there will be a lot of ego death, esp. with people deep in a lot of kowledge work.
But if you claim that AI is absolutely a waste and bad for humanity I feel sorry for you if you can't see how incredible it is that we have this new technology. -That it will most likely solve some very core issues humans are working on, and that it is a kind of evolutionary moment in human history, creating artificial intelligence. I feel sorry that you can't see past your own nose to see the bigger picture in that sense.
I tend to think big changes will always make things better and worse at the same time. If you look at the long horizon of history (100s of years) it appears that it makes things better on balance. But who really knows for sure what will happen.
I think at this point I am mostly in AI camp. The recent advances in reverse engineering automation ( and subsequent panic -- and upcoming banhammer - coming from owners and shareholders of previously available money makers ) made me recognize it is a real shift and not just an incremental improvement.
People are scared. I suppose they should be. A lot of people have been invested in ensuring that things stay roughly the way they are. I am not sure that was ever sustainable, but now it seems like an impossible dream.
Cat is out of the bag. Milk is spilt. Hanninal crossed the alps.
Am I worried? Yes. But it is still fucking amazing to watch it unfold.
This will solve problems that humans are working on, but it also will solve everything that makes life worthwhile.
There's nothing good at the end of this path.
> it also will solve everything that makes life worthwhile.
It's going to solve your kids? Or eating food? Or enjoying fresh air?
> will solve everything that makes life worthwhile
Like programming? Or what else, I'm not sure I understand what that would mean.
Like problem solving, art, music, writing, and other intellectual challenges. And doing it well enough that you make an impact on the people around you.
Programming is one such creative endeavor, but they're all under attack.
Personally, when it comes to media and creative things, I want to hear from other humans. It could be other humans using AI as a tool, but I'm not interested in the Meta AI account slop they were promoting. Or the Spotify algorithm optimized playlist artists.
I don't think everyone will agree with me, and we might have truly AI generated movie and pop stars, but I think most people won't want synthetic art- it's missing the point of what art is for, for expressing something about humans.
Right now it's easy to say, because the AI generated stuff is so obviously bad. What if it were better than human output? What if the only way to tell was that the human sounded worse?
You're betting on AI continuing to be bad, or on humans rejecting it wholesale. While I wish humans would reject it, I'm not so optimistic.
This "organic" comment is from an AI booster:
https://x.com/awongh
The AI generated posters are an utter disaster. What a time to be alive! Shills, poseurs and influencers.
My personal emotional reaction to these changes is very similar, as a 40-something year old programmer. I wrote my own take on it (https://medium.com/@kannanvijayan/talking-to-the-computer-5c...), didn't submit it to HN though.
I was genuinely considering whether or not I still liked programming - the excitement and risk of technical failure had dropped off. Most failures were for other organizational and process and management reasons. I respect the value of executing those aspects well, but I have no interest in becoming fluent in solving that class of problems. It just bores me. I want to build big beautiful, complex, intricate machines.
The agentic tools are, for me, a step up into tackling broader projects I had put on the shelf as "not doable without investment". I can self-fund my ideas now. Things I anticipate would have taken about a decade or so to play around with and build out as a sole developer, I'm taking on as a sole developer.
But that said, I do appreciate that we can't be sure of what exactly this churn will produce. The other factors that go along with the technical achievement aren't too rosy.
<< I can self-fund my ideas now. Things I anticipate would have taken about a decade or so to play around with and build out as a sole developer, I'm taking on as a sole developer.
This. All of a sudden, I have real personal goal timeline of things I want to build. The odd thing is, there are still hard things on that horizon, where I can see clearly things will require a team of humans, who understand how things really work under all these abstractions.
Hmm, sometimes I feel like living in a different universe. Since LLMs can solve any technical challenge (like most of you claim here) and help folks explore new fields and revive all sorts of projects, wouldn't be all seeing effect of that everywhere? Wouldn't the bar for software quality go up (and go up significantly)? Wouldn't we see new innovative apps that people are delighted to use and more than happy to pay money for? Yet when I recently challenged some friends of mine to name just ONE major app / service that became materially better (I'd argue even better at all) or a new/innovative widely adopted by masses outside of pure AI world (Claudes and GPTs), they couldn't name any. Meanwhile, I am looking around and see that banking, insurance, search, social media, cloud computing are getting more bloated, unstable, mediocre and in general of lower quality. Where is all the cool software everybody is talking about? Where are the features from your backlogs that you were finally able to burn through? I am just not seeing it in the real world. Something doesn't add up here.
In the past we had communities. They provided us with a context of meaning for our lives. We have lost that. So the context people have for their lives is either money, or just themselves.
Many people have not even experienced community these days. Imagine where you feel part of something bigger than yourself. A gestalt, where the whole is greater than the sum.
We People grew being in that situation - we would hunt, farm, fix things, make things - better not just for ourselves but for our community. It became our nature, just as it has became the nature of sheep dogs to herd.
Imagine a sheep dog who lives a life with no sheep. They have all that built into their nature and never get to be that. That is you and so many other people. So there is your challenge.
It is interesting that you say that, because it may explain my relative ease with this particular transition. Community ( as understood these days -- not exactly a neighborhood, but not just your household either ) was not really my guiding light before so its potential loss is not as pronounced to me. In other words, loss of that context does not visibly affect me.
> P.S. I am a human and can make mistakes. The spelling and punctuation in this post were corrected with AI.
There's a challenge for you right there. Learn to spell or learn to live with mistakes.
I categorize spelling and grammar corrections as basic fact checking. I like to write how I talk and that means I sometimes leave the bad grammar because it seems more like my voice (whatever that is).
And it would be bad if the article used AI for basic fact checking too.
Not actually worse than for spelling, because both use cases equally allow it to insert bullshit into the text.
I learned to live with my mistakes :)
I remember the old USENET Days, back them people were noto concerned with spelling, at least in the groups I read.
That's really dumb challenge in my opinion.
Writing code is arguably completely useless and dumb too
did u have a way to back that belief ?
Isn't it the point of the post though? Why challenge yourself to learn how to spell if AI can just fix it for you in 2 seconds?
Why learn anything if AI can do it for you?
I think that’s the whole point the article/comments are making.
Many things we used to learn are becoming obsolete by AI.
A analogy (if not the best one)3 People used to be much less sedentary, and in some ways healthier, before automobiles (or working from home).
So setting aside intentional exercise became much more necessary to be less sedentary and healthier.
But it requires discipline and motivation, and many people don’t exercise enough as a result.
Before AI, learning was a necessity to get things done. We had to check our own spelling/grammar before these tools existed. Etc.
Now, learning and using these skills will be simply for the “exercise” of using our brains in ways that were necessary before AI.
Some people will continue to code ok the side “for fun” or “for exercise”, but it will no longer be a means to simply getting something done — that can be done with AI.
Any you certainly won’t be doing for a company whose primary focus is to make money.
Yes. This is the question kids are also asking themselves as the discussion of AI companies solving white collar labor filters down.
A significant portion are deciding that the game is rigged. I don't know if I could look them in the eyes and tell them them that they're wrong.
Should either be
> ...spelling and punctuation mistakes in this post were corrected with AI.
or
> ...spelling and punctuation in this post was corrected with AI.
So it highlights the point.
Spelling and punctuation are two different things, are they not?
This might be the end of personal challenges but we have so many really hard challenges that could do well with a lot of people coming together. How do we solve for poverty and the lack of clean air, water, and food in urban India (could be anywhere, but this I see daily). What could we do to get people to collaborate at scale on problems that are not currently profitable for an org? Cheap air conditioning. Senior citizen loneliness. So many problems to take up. We probably can't even think in that direction because these seem so huge now. But that's hopefully what intelligence on tap is now bringing into the tractable domain
I've tried a few things that I know are possible (multiplayer FPS with client-side prediction), but the overall challenge still persists. There is no easy way to rapidly iterate this architecture with an agent unless you already have massively deep pools of experience to draw from. Launching this system into the real world is even more challenging than building it.
For many problems, the LLMs look like they might be making progress, but once you zoom out far enough you realize it's just pushing the toothpaste around in the same tube with the cap still on. The mass of the overall system is not changing despite the appearance of localized progress.
If your primary source of identity was gatekeeping the Pantone colors of the various office furniture, then yes perhaps the challenge is now gone. But, if you were focused on mission and 'big picture' (what furniture and why), it's plausible that the challenge could actually increase in its intensity.
I don’t know man, a lot of shit I try still doesn’t work. Maybe the challenge has relocated from doing the right thing to asking the right questions and knowing what the right answer looks like — but it’s definitely not “gone”.
It works well enough that the work remaining is the sloggy manual testing and bug reporting. You're right that it doesn't work well, but the interesting part has been automated.
This is true in general: janitorial work is likely to outlive white collar work in this hell we're building.
Not "gone". But definitely "going". And fast now.
It takes only a few seconds to discover dozens of unsolved engineering challenges in multiple fields. The only things that are going are the trivial ones. Ignoring that, LLMs bring a whole slate of their own challenges.
> LLMs bring a whole slate of their own challenges.
Yeah but will it pay off to invest time in those challenges and become an expert in them? My gut feeling says no.
Challenge is not gone. Use LLMs to clear up your backlog of things you wanted to do before but couldn't, then choose something that's challenging even with the help of an LLM. Dive into a skill you thought you didn't have time to learn. Make a something you previously thought was too big to make by yourself.
I’ve been trying to get better at analyzing performance traces (like Bruce Dawson has written about on his blog), crash dumps, and assembly language in general.
AIs are still pretty bad at all this stuff.
It’s made we wonder if future AIs will get much better at machine language and will skip writing human readable source code and just dump out CPU opcodes (ie write the binary directly).
What I want to do is think about the challenges in my backlog.
And help the AI labs to become good at it ...
This. Running a load of things on a server at home was on my backlog for ages. I started manually then got Claude to reverse engineer the system and create a load of Ansible that we iterated on. It still needs its hand holding , and likes to pretend it doesn't know what good Ansible looks like but we got there
It also wrote a nifty remote LUKS unlock utility. So handy
I think the author (and many people, including myself) is missing something crucial.
Yes AI can solve very difficult problems without requiring you to understand enough about it.
This gives the impression that you can solve anything, create anything, and put it out there. And because this seems to be true, you feel less motivation to go and do it, because it’s so easily achievable. It seems AI is robbing your ability to differentiate yourself and push hard on a direction and build something that matters.
However, I have a feeling that is increasing over time. I believe that this is not true at all.
Even with AI, an enormous amount of energy needs to come from within one self, to truly create something meaningful that is used by other people. Just because AI can spit out code that solves a set of problems, it doesn’t mean it’s ready to be used and that it will turn out to be something people actually care and use.
If your only ambition was solve a hard problem, and the problem was purely technical, I agree you are in a bad spot, but then again this is not gone. People play single player videogames fully knowing they can finish the game and they can even hack if they wanted but they don’t, they still enjoy playing, it just means this is now perhaps a hobby rather than something you can profit from.
Creating a product however, remains difficult I believe. A product that people actually enjoy using and solves something. If anything the bar is higher, and it is very hard to differentiate. I think this is more challenging than before, not less.
So if you shift your ambition to be, I want to create something people actually enjoy and it’s useful, and I can survive the next few years and have a competitive advantage, I think this is incredibly difficult.
I do miss programming instead of reading code. But I enjoy designing and picking languages and tools. For me AI has been great (so far): I revived old projects and came up with new ones, including branching to hardware products. This would have never been achievable without AI. Not because a lack of skill. I'm sure I could have picked any of the projects and see it through. But I wouldn't have, because the time and effort wouldn't have justified the result. And I would never have been able to do so many of them. I would have needed to pick one, pour many, many hours of work into it and all the other things would need to wait. So I could have done a fraction, but in reality I would have hardly done any. So for me it's enabling.
It's not that I couldn't have. I wouldn't have.
Currently, for me the challenge has gone from can I build it, to can I build this and everything else I need, and structure it to be maintainable for the long term, so I can keep stacking bricks. This has found me leaning harder into IaC, use of and contribution to open source (for secure ownership, longevity of system and cost reduction) and structuring the development tooling around agents. That will be an ongoing process, but I feel that I'm at the point where I can move onto building what matters. I've already built a few of the tools I've dreamed of having for years (there's a couple of rough edges but it beats the tools I had before), and I think, like the author this will free me up to stay on the next step (building what matters) for good.
I think there's still challenges.
The main one I see is the question of how we can use an LLM as humans to produce something harder than the LLM can by itself.
I like to think of it with an analogy. Suppose I was granted the privilege of having a world renowned human expert as my personal assistant, but she had no initiative of her own. How could I collaborate with her?
I would definitely need to keep learning more to understand what she says. I would also need tools to augment my ability to follow along and she would need tools to bring knowledge down to my level of understanding.
One might argue humans are unnecessary in the above process, but I think that's just failing the challenge of keeping up. So I try
We've largely reached the point where a mathematician in the loop is a net negative in theorem proving.
How long do you think other fields have?
That's what I'm saying though. It's a new challenge to make the mathematician a net positive again.
We can consider what makes a human a net negative in the loop and augment humans with tools to collaborate.
I don't know, the set of problems that remain very hard but worthwhile to solve still massively dwarfs the set of problems that has become easy to solve with the help of AI, even in a restricted domain like software engineering.
It is a failure of imagination to think that there is no longer much challenge. But I guess all humans have a problem with tunnel vision, to various extents.
Come join those of us that enjoy solving problems at work but aren’t driven by it. The water is fine, the weather is usually decent, and the existential terror of “failing” goes away eventually. We have the same problems and worries but sleep a bit better when things go wrong.
Ugh. Boring as fuck.
All challenge is gone? No way. Your next challenge is to find a new one.
LLMs make what was formerly hard trivial, that is true. But they cannot do everything with simple prompts, you must increase your ambition to what was formerly impossible. Try building a bigger game, more complicated software or hardware…what have you always wanted but written off as something only a multi-millionaire with a large team of domain experts could accomplish? Because you have one now.
My initial reaction as well. If you believe AI has made what you do irrelevant, change what you do. If you really want to do something different, stop playing the knowledge arms race game with a computer program. There has been "analog" LLM-like ventures (aka big, rich tech corporations) for a long time, yet they didn't previously solve all engineering problems; you can now use technology as a great equalizer, or change your focus to something AI can't even do if you don't want to play the game.
I was glad to see the direction this essay took. I agree with you: it's the old challenges that are gone and there's a whole new class of challenges (the former impossibilities) in front of us.
That's a feeling I've been also feeling for the last year or so. Turns out the means was what I cared about, and I didn't care about the ends that much. LLMs robbed me of that in a way I couldn't see coming.
I think that we as a society fail to see that the speed we built things was rarely the problem. The real shortage was in the ideas - having them, finding the motivation to act on them, sharing them, and the biggest driver was the desire of showing off, being told you are great and your idea was great.
I am afraid that by removing the ick to fix things, to challenge everything and being rewarded for that, we will end up with less innovation, not more. Why bother? Why do anything if everything is cheap, why start a new project only to be shrugged away as being yet another thing any machine could have done? Why show off when the merits are the machine's?
HN went from being a constant stream of Show HN to AI news after AI news. The most interesting projects AI have enabled are to run old stuff onto other things.
Rehashes of old ideas keep sprouting out of everywhere, keeping new ones at bay. I think we are probably entering a long winter and all that will be left is permutations of what we did before '23
Your next challenge is to finds a next challenge.
Ai helps me build better, more deterministic outcomes that it itself can't. I mean I'm building a mailing list manager that blends names of couples. Yes, AI can do this but it's inconsistent. I can direct AI to code precise outcomes step-by-step. Coding is solved? Not sure, but my directing code is solved.
Challenge of chasing food is gone, but I still walked 12km this morning and feel great right now. Apply this approach to everything else in your life.
> I was only ever interested in engineering challenges.
When you pigeonhole yourself into only finding challenge in one field, you're bound to be disappointed eventually. While I do miss some of the challenge, these tools have given me more freedom to tackle difficult problems outside of software, both personally and professionally.
Yesterday I turned 40 and reflected on the same thing.
For most of my life the “main quest” was always obvious. I went to school, found my soul mate, got married, bought a house, had kids, paid off the mortgage, got yelled at by Howie Mandel, won the awards, achieved my career goals.
But now there’s no main quest. It’s just a whole wonderful, brilliantly rendered world of side quests for me to pick from. I’m spoiled with choice…
The thing is… main quests drive you, but you have to drive the side quests; they’re perfectly content to not get done. It’s an adjustment, and I’m optimistic that I’ll overcome it. But I’m currently trapped in the doldrums of figuring out what to do between now and when the credits roll.
We've clocked the game and now just waiting for someone to turn off the console?
You write well, why not challenge yourself to write a book?
Assuming you didn't use AI for more than spelling
Who/what is Vestigius Terrentius?
LLMs might help you find harder challenges.
I can solve any engineering problem (that I can come up with) with the help of artificial intelligence.
Great. Practical nuclear fusion power, please.
The author says he vibe-coded a book translation website, and there is no challenge.
Anyone who has translated a book, a literature one, knows it isn't something that is well done by AI.
Indeed; but for the author, the challenge was not to produce a good translation of anything, or to develop a good automated process for doing so. It was to bash together a website that churns out translations, without even a specified level of quality. For the author, these are all challenges of making a process to do something; not to produce any actual artifacts of quality.
It's easier to believe you can solve any engineering problem if you limit your thinking to problems you comfortably believe you can solve..
I feel the opposite.
I feel like possibility has arrived.
I can build anything I want.
Even so it’s still very challenging to build anything worthwhile.
Challenge is never gone.
Here's a challenge: see yourself from the outside.
Realize that however effective you are as a programmer now, you were once a hopeless beginner.
Then you must acknowledge that you are still a hopeless beginner, relative to the you that doesn't exist yet.
In the spirit of fellowship, I tell you:
---And as you continue to learn, you'll see this for yourself, but I guarantee to you now that AI opened paths to writing novel software which no one has noticed yet. Software that was possible before but perhaps annoying to build, etc. Software which perhaps will only ever be imagined by someone who learned exactly the things you have, in exactly the order you did.
Never make the mistake that AI knows how to do everything just because it knows how to do everything we've done so far.
And don't think of your work as "mining" engineering problems! Forget engineering problems. Find a human problem and think about it for an entire month.
understanding of the unconscious would be entirely novel , yet such knowledge is overtly claimed here. if the author is human he may be showing signs of ai psycosis.
Being a SWE is now a mix of being a PM and reviewing vibecoded slop all day
I'm so glad I chose a different path!
> I simply know that I can solve any engineering problem (that I can come up with) with the help of artificial intelligence.
I'm sorry, but I find this a bit silly. Either:
* your imagination is extremely limited (which is OK, no worries!), and the statement is true because of that, or
* your imagination is fine and the statement is just false, or
* your imagination is fine and the statement is true, in which case: what's stopping you? We all agree that there are engineering challenges, the solution to which would make the world a better place (or make you fabulously wealthy, whatever floats your boat), so get cracking if it's so easy now!
Which is it?
[flagged]