I've noticed a huge drop in negative comments on HN when discussing LLMs in the last 1-2 months.
All the LLM coded projects I've seen shared so far[1] have been tech toys though. I've watched things pop up on my twitter feed (usually games related), then quietly go off air before reaching a gold release (I manually keep up to date with what I've found, so it's not the algorithm).
I find this all very interesting: LLMs dont change the fundamental drives needed to build successful products. I feel like I'm observing the TikTokification of software development. I dont know why people aren't finishing. Maybe they stop when the "real work" kicks in. Or maybe they hit the limits of what LLMs can do (so far). Maybe they jump to the next idea to keep chasing the rush.
Acquiring context requires real work, and I dont see a way forward to automating that away. And to be clear, context is human needs; i.e. the reasons why someone will use your product. In the game development world, it's very difficult to overstate how much work needs to be done to create a smooth, enjoyable experience for the player.
While anyone may be able to create a suite of apps in a weekend, I think very few of them will have the patience and time to maintain them (just like software development before LLMs! i.e. Linux, open source software, etc.).
[1] yes, selection bias. There are A LOT of AI devs just marketing their LLMs. Also it's DEFINITELY too early to be certain. Take everything Im saying with a one pound grain of salt.
> I've noticed a huge drop in negative comments on HN when discussing LLMs in the last 1-2 months.
real people get fed up of debating the same tired "omg new model 1000x better now" posts/comments from the astroturfers, the shills and their bots each time OpenAI shits out a new model
Simply this ^ I'm tired of debating bots and people paid to grow the hype, so I won't anymore I'll just work and look for the hype passing by from a distance. In the meanwhile I'll keep waiting for people making actual products with LLMs that will kill old generation products like windows, excel, teams, gmail etc that will replace slop with great ui/ux and push really performant apps
Especially when 90% of these articles are based on personal, anecdotally evidence and keep repeating the same points without offering anything new.
If these articles actually provide quantitative results in a study done across an organization and provide concrete suggestions like what Google did a while ago, that would be refreshing and useful.
(Yes, this very article has strong "shill" vibes and fits the patterns above)
You're only hurting yourself if you decide there's some wild conspiracy afoot here to pay shills to tell people that coding agents are useful... as opposed to people finding them useful enough to want to tell other people about it.
If I worked for Microsoft as a software engineer and believed that LLMs were going to end software engineering I would not expect the value increase in my stock options to overcome my loss of income when Microsoft inevitably laid me off.
(I do not think LLMs will obsolete software engineering as a career.)
I don't assume everyone can think of that next step.
If I were that smart, I would not be writing a blog article just talking about using LLM to create new projects/tools outside production environment, because the same thing has been written 1000 times at least, and this article would not offer anything new, which would be a waste of time.
Which unfortunately is what's happening here.
(I came to HN comments of this article to look for new perspectives. I found exactly nothing.)
Why is it the people posting positive comments who are "responding to incentives" by posting more, while it's the people posting negative comments who do so by stopping posting? Like, your exact points work equally well with the polarity reversed: the anti-AI influencer/grifter ecosystem is well-developed at this point, and many people desperately want AIs to be useless.
I don't know if the original claim about sentiment is true, but if it is, I don't think yours or blibble's (conflicting) claims about the reason are very believable.
> Like, your exact points work equally well with the polarity reversed: the anti-AI influencer/grifter ecosystem is well-developed at this point, and many people desperately want AIs to be useless
Maybe it's equal for non-tech people. But I don't think a lot of tech people are desperate for AI to be useless, I think they're desperate for it to be useful.
If you're someone who is smart enough to work with or without AI and you just find the tools not that helpful, I doubt you're all that worried about being replaced. But when we see companies increasingly bullish on something we know doesn't work that well, it's a bit worrying.
because there's no sweet tech-oligarch job, early access to the latest model, OpenAI speaking engagement invite, or larger bonus to be awarded by being aiphobic?
This is a cringe comment from an era of when "Micro$oft" was hip and reads like you are a fanboi for Anthropic/Google foaming at the mouth.
Would be far more useful if you provided actual verifiable information and dropped the cringe memes. Can't take seriously someone using "Microslop" in a sentence".
It could be that the people who are focused on building monetizable products with LLMs don't feel the need to share what they are doing - they're too busy quietly getting on with building and marketing their products.
Sharing how you're using these tools is quite a lot of work!
I admit I'm in this boat. I get immense value from LLMs, easily 5x if not more, and the codebases I work in are large, mature and complex. But providing "receipts" as the kids call it these days would be a huge undertaking, with not a lot of upside. In fact, the downsides are considerable. Aside from the time investment, I have no interest in arguing with people about whether what I work on is just CRUD (it's not) or that the problems I work on are not novel (who cares, your product either provides value for your users or it does not).
Agreed! LLMs are a force multiplier for real products too. They're going to augment people who are willing to do the real work.
But, Im also wondering if LLMs are going to create a new generation of software dev "brain rot" (to use the colloquial term), similar to short form videos.
I should mention in the gamedev world, it's quite common share because sharing is marketing, hence my perspective.
I feel weird when I read comments that have words like "force mulitplier". This sounds like an LLM comment. But you probably are a real person. So are you just becoming more like an LLM because you interact with it so much, or did you always talk like this and LLMs are just replicating that behavior?
If you haven't noticed, people come here to kill time. If you're killing time then you're not being productive, therefor the people who are heads down trying to launch their startup before they run out of runway are not going to be here until they're ready to market their product.
The type of people to use AI are necessarily the people who will struggle most when it comes time to do the last essential 20% of the work that AI can't do. Once thinking is required to bring all the parts into a whole, the person who gives over their thinking skills to AI will not be equipped to do the work, either because they never had the capacity to begin with or because AI has smoothed out the ripples of their brain. I say this from experience.
I think you can tell from some answers here that people talk to these models a lot and adapt their language structure :( Means they stop asking themselves whether it makes any sense what they ask the model for. It does not turn middle management into developers it turns developers into middle managers that just shout louder or replace a critical mind with another yesman or the next super best model that finally brings their genius ideas to life. Then well they get to the same wall of having to learn for themselves to reach gold and ofc that's an insult to any manager. Whoever cannot do the insane job has to be wrong, never the one asking for insanity.
Sad i had to scroll so far down to get some fitting description of why those projects all die. Maybe it's not just me leaving all social networks even HN because well you may not talk to 100% bots but you sure talk to 90% of people that talk to models a lot instead of using them as a tool.
My thinking is definitely better. I spend more time worrying about the specific architecture, memory layout, GPU features, etc. to come up with ideas for optimisations, and I think less about specific implementation details. I’ve gotten a better mental model of our code faster because of this. I have also found substantial speed ups by thinking about the problem at a higher level, while iterating on implementation details quickly using Opus.
Deploying and maintaining something in a production-ready environment is a huge amount of work. It's not surprising that most people give up once they have a tech demo, especially if they're not interested in spending a ton of time maintaining these projects. Last year Karpathy posted about a similar experience, where he quickly vibe coded some tools only to realize that deploying it would take far more effort than he originally anticipated.
I think it's also rewarding to just be able to build something for yourself, and one benefit of scratching your own itch is that you don't have to go through the full effort of making something "production ready". You can just build something that's tailed specifically to the problem you're trying to solve without worrying about edge cases.
Yeah, I do a lot of hobby game making and the 80/20 rule definitely applies. Your game will be "done" in 20% of the time it takes to create a polished product ready for mass consumption.
Stopping there is just fine if you're doing it as a hobby. I love to do this to test out isolated ideas. I have dozens of RPGs in this state, just to play around with different design concepts from technical to gameplay.
Sometimes I feel like a lot of those posts are instances of Kent Brockman:
"I for one, welcome our new insect overlords."
Given the enthusiasm of our ruling class towards automating software development work, it may make sense for a software engineer to publicly signal how much onboard as a professional they are with it.
But, I've seen stranger stuff throughout my professional life: I still remember people enthusiastically defending EJB 2.1 and xdoclet as perfectly fine ways of writing software.
Not mine, but the reason I find this interesting is how long it took to reach that point.
4 months! Each LLM improvement pushes this further and further out. Depending on the complexity of the project, there'll be a point where 4 months turns into 6, 8, 12, etc months until the code generated by the LLM is too large for it to manage. Lots of interesting implications.
This year I've been diving into Game dev books and Harry Potter fan fiction. There's a lot of great reads out there. I assume having an entire world built for you removes a lot of the heavy lifting (as a writer).
GameDev:
- The Masters of Doom
- The Doom Guy
- Play Nice
- Press Reset
- Blood, Sweat, and Pixels
- Console Wars
- Ask Iwata
- Embed with Games
HP Fanfic:
- All The Young Dudes (by MsKingBean89) [student life of the Marauders]
- Grey Space (by noaacat) [Pre Hogwarts. Focuses on parental abuse. One scene I'll never forget (paraphrasing) is when another student at school notices Harry's bruises, slides up her sleeve, and implies she slipped too]
- Just started the Glasslight series (noaacat).
I also reread the entire Harry Potter series, and revisited a handful of Redwall books. I find it interesting how I loved RW as a kid/teen, was bored out of my mind with it in my 20s, and now I love it again.
Every time I think about EVs, I become filled with dread thinking about what will happen if an area loses power for an extended period of time. I used to live in an area that had transformers blow from rain showers.
At least you could hook up a generator to pump gas at a gas station.
:/ Life's about trade offs.
Edit: Whoops.. Im not against EVs to be clear. But from a safety POV, having two different energy sources is safer than having one. Im not sure if you'll understand this if you haven't lived in a very snowy state.
You could also flip that and talk about the risks of when your gasoline supply get shut down due to some event. With an EV stack you can generate your power locally and add resilience that way.
my area lost power for about 3 days last week and I ran all of my house's critical systems from my EV. it was great - silent, unlike the old generator, and not counting the sunk cost of the car, extremely cheap. Cost maybe $5 in electricity to keep the furnace, refrigerator/freezer, and internet on for 3 days, contrasted with probably $50 in gasoline for a similar amount of time.
If the outage had been longer, I could have made a half-hour trip to an area that had working EV fast chargers and come back with another 5-6 days of power for the house.
How expensive is the setup to tie in your vehicle into a house grid? Or can it be something as crude as running extension cables to plug into a powerstrip attached to the car?
I did it cheap - I just ran extension cables from my car (previously from a generator). I have a manual transfer switch at my furnace to safely switch between utility power and extension cable power, and just replugged appliances. I think everything drew on average about 800 watts, up to maybe 1500. I split it between 2 1.8 kW 110v outlets in the car.
If your vehicle has V2H (Vehicle to Home) the way it generally works in newer cars is that you have a bidirectional charger that can be used to connect the car battery to an external inverter, which is then tied into your house grid the same way a generator would be, e.g., via a transfer switch on the circuit you want to power.
This isn't cheap--I think it was generally many thousands of dollars for the cars I looked at that had V2H when I was EV shopping--but it generally gives you something something in the 9-20 kW range of power. For most people that gets into the territory of "as long as you remember not to use the electric clothes dryer at the same time you are making a meal that uses the electric oven and all of its burners you can just continue as if power is not out".
I seriously considered it. In the summer I use around 8-10 kWh per day, and in most of winter under 40 kWh per day. With a car with an 85 kWh battery, keeping it in the 20-80% range, and keeping it near 80% when outages are likely, That would give me several days of backup power during a summer outage, and over a day during a winter outage. In 18 years in my current house I think I've only had one outage that went over a day, and never had an outage so widespread that I would not have been able to find a public DC charging station within 10 miles to recharge the car if an outage actually did last long enough to take it down to 20%.
But the cars with V2H were out of my price range, even before adding the cost of the equipment to use V2H.
If your vehicle has V2L (which is what the one I bought has) it is considerably cheaper. The car gives you one or two outlets similar to ordinary household outlets. Mine gives a single outlet, which you get by plugging an adapter into the charge port. With these you generally don't try to tie it into your house grid. You just run an extension cord (or two if your V2L provides two outlets) to where you want power.
Some people get some sort of socket installed on the outside of their house that a cable from the car's outlet can be connected to, with that socket connected to an indoor outlet. Me, I just leave sliding door open enough for an extension cord to go through, and then stuff the gap with some foam strips that I got at Home Depot.
It is surprising how much a single 120V 15A circuit can do. What I need to get through a one day power outage comfortably (which as mentioned would be an unusually long outage here) is: (1) power for the fridge, (2) power for an electric space heater, (3) power for my computer area and cable gateway (if cable is not out), (4) maybe power for some cooking, and (5) water to flush the toilet.
Any time the weather forecast even hints at something that could cause widespread outages I fill a bathtub with water for #5. For the rest I've been monitoring power used for those things with a bunch of energy monitoring smart outlets (Tapo P110M controlled by Home Assistant using Matter).
For #1, my fridge during a normal cycles draws 90-100W. During a defrost cycles it draws 400W. I have wireless thermometers in the fridge and freezer compartments so I can easily coordinate with other uses such as cooking to make sure I cook at a time when the fridge is going to not need to run for a while (which I can ensure by unplugging it).
For #2 my space heater is usually 1500W, but I've got another one that is supposed to be 1500W but due to age is only 1300W, and I've got year another one which is 1500W on high but has a medium setting that is 1000W. On all but the coldest days 1300W and probably 1000W would keep it warm enough as long as I'm warmly dressed.
For #3 my entire computer setup (Mac Studio, 27" 5K monitor, 24" 1920x1200 monitor, speakers, external Thunderbolt drive bay with 4 SSDs), a network switch, and a Hue hub is about 130W with short spikes to around 170W. If I turn off the second monitor that drops about 25W from that. The cable gateway is 15W.
For #4 I've got a microwave, a toaster oven, and a couple George Foreman grills. The microwave draws over 1900W for the first minute or so on high, but I can set it lower and it is an inverter microwave so on lower settings it actually reduces the amps drawn rather than just cycling between full and off. The toaster oven is 1000W and the biggest GF grill is 1200W.
I should be able to run all of these, as long as I take some care to not run too many at once. Some observations:
1. If I'm not going to use the computer for a while, such as when sleeping, I can run the 1500W space heater, turning it off when the fridge needs to run. I could actually then turn it back on once the fridge has started and gotten past its inrush current (20A, which my V2L has no trouble with). It would then be 1600W total unless the fridge is doing a defrost cycle than it would be 1900W. That's fine because the 400W defrost phase only lasts about 10 minutes and the non-defrost cycle only a bit over an hour, and it is around 3 to 8 hours (depending on how often I open the fridge I assume) after a cycle ends that it needs to run again. That counts as an intermittent load and so should be OK with my extension cords (rated 15A intermittent, 12A continuous).
2. If I'm using the computer I can switch to the 1300W space heater. That plus the computer both in continuous used would be under the 12A continuous rating of my extension cord. When the fridge needs to run I'd have to switch to a lower setting on the space heater until the fridge is done.
3. When I need to cook I'd just need to time it so it happens when the fridge won't need to run, and turn off the space heater while cooking. Nothing I'd be cooking in the toaster over or the GF grill takes more than 20 minutes. The computer stuff could remain powered during this.
4. If I don't have a bathtub full of toilet flushing water, I might be able to run my well pump from V2L. It's 120V with a 1/2 HP motor which would be under 1000W, but the inrush current may be too high. The specs say maximum of 44A, but from what I've read many people have had success with motors with that kind of inrush current, and the way the V2L system works it is safe to try it--worst case is the V2L safety systems shut it down. I just haven't gotten around to trying it yet.
Overall then it seems like that single 120V 15A from V2L will actually be enough to get me comfortably through most power outages. I had not expected that when I got the car.
That sounds like the extreme version of "but I need a fuel car because I want to drive it to France once a year for holiday". Driving something around all year for a once-a-year event is silly, but this is just insane. In a good life, you don't need this fallback from grid power even once in your lifetime!
At least, not beyond the inconvenience that is having to stay at home like 1 unplanned day per several decades. That's still three and three quarters of a nine of uptime even if you'd get the recent Iberian peninsula event every 10 years, and assumes you emptied the battery coincidentally the day before the outage. If you're not an EMT or power plant technician, you're doing more harm than good by being the person who can drive to work during a power outage and find that you're the only one there and nothing works anyway
> That sounds like the extreme version of "but I need a fuel car because I want to drive it to France once a year for holiday".
Having just made the 1,000km trip to the French Alps and back again in a Tesla Y, that's not a valid excuse any more. Back at home in Australia driving 2,200km from Mackay to Melbourne in a EV also a common enough holiday trip.
The 5,000km trip to Perth might be a stretch, but it's considered a major undertaking in a conventional car too. You are crossing some of the most remote places on the planet that has paved roads. The problem isn't charging. It's that you need to carry spares - like drinking water for emergencies, and spare tyres.
It's the tyres that would stop me from doing it in a Tesla Y. The Y doesn't have a place for a spare tyre, which is a disease that seems to afflict many modern cars of all types. It doesn't even come with a jack. Worse it needs special tyres that are hard(ish) to find in a major city, let alone 1000km from anywhere.
Unless grandma lives in a place without electricity, the one issue you won't have in Australia or Europe is charging. EV charging points are everywhere now. Most parking lots have them. I dunno what the situation is in the USA, but if EV charging points are a problem I'd suspect deliberate government interference because in Australia at least every one seems to have been built privately. Unlike Europe Australia does not have much in the way of EV subsidies, yet they are springing up like weeds.
I suspect the reason is location, location, location. Similar to petrol stations, but unlike a petrol station the upfront investment is low, they aren't manned so no wage costs, in a shopping centre they attract customers and they seem to markup the cost by 80..150%. What's not to like? So get in early and get the best spots.
I guess I didn't make it clear that this quoted argument was meant to sound silly. I know you can go on holiday with EVs! My mom was a bit apprehensive about spontaneously needing to find chargers and figure them out, and so I invited her to a weekend trip for just the two of us, one of the goals being to see together how charging works out in different places and countries and how often that ends up being necessary etc. It was no problem at all if you can figure out the different payment methods (most worked with a magic card she got with the car, others wanted paypal via a web portal etc.)
Considering how expensive cars are, I do find the trip-to-grandma reasoning useful. Most people want a single vehicle that can do everything. Dismissing that with, "Well just drive differently" or "You can do the hassle that is renting a car" is not a compelling sell. What if I want to do my vacation trips during the holidays where rentals are already booked?
I think full EVs are great if the lifestyle allows it, but plug-in hybrids seem a better fit for most people without requiring undue compromise.
What trip to grandma can't you do anymore with an EV?
> What if I want to do my vacation trips during the holidays where rentals are already booked?
The same as you do when any other part is booked out: go elsewhere or do something else. I don't buy a backup train in case the one I want is booked out one of the next ten summers
Consider also the lifestyle change that's "growing older more healthily" by not having a population sit in exhaust fumes for 2x the daily average commute length
Why there focus on sitting in traffic. Instead of more visionary solutions like banning single family homes and razing them all to ground and replacing them with high rises next to offices supported by forced public transport? Surely that should be the true alternatives for use of ICE and not EVs.
We’ve already had some experiences with this in Florida after a hurricane. EVs do quite well, turns out, better in many cases than gas cars. The grid is a very high priority for fixing, more so even than gas supplies.
> Every time I think about EVs, I become filled with dread thinking about what will happen if an area loses power for an extended period of time
A mid-size EV battery can easily store 60kWh of energy. That's enough to power a domestic refrigerator for ~12 days (assuming 200w average power usage, which is on the higher end).
I lost power for about a day last year and was very happy to be able to keep my fridge and some emergency lighting powered from an EV battery.
Rooftop solar is getting cheaper, easier, more efficient, and more reliable every year.
If you live in an area that is poor enough that this is not an option, it loses power frequently due to weather, and no one in power cares enough to fix it, that genuinely sucks, and I feel for you. But, as sibling comments said, some other poor areas don't get gasoline shipments in a timely manner—being poor and neglected is just always going to suck in various ways, and the solution is not to avoid any technological advancements that remove the crutch that your particular poor and neglected area is using to get through it a little easier, but to find ways to reduce the poverty and neglect.
And, frankly, solar power and electric vehicles are both great tools to help with that, especially when used together.
The common advice among "preppers" and more anxious individuals is to never let your gas gauge go below half or 1/4 tank. I think people are much more anxious than they realize about running out of gas in a black swan event.
When a hurricane impacted my area last year, I kept seeing Facebook posts for a week or two afterward from people asking where to find gas. Meanwhile, the power never went out, so my EV was able to charge without issue.
Over time, do you think this process could lock you into an inflexible state?
I'm reminded of the trade off between automation and manual work. Automation crystalizes process, and thus the system as a whole loses it's ability to adapt in a dynamic environment.
Nothing about this feels inflexible to me at the moment - I'm evolving the way I use these tools on a daily basis, constantly discovering new tricks that work.
Just this morning I found out that I can tell Claude Code how to use my shot-scraper CLI tool to debug JavaScript and it will start doing exactly that:
you can run javascript against the page using:
shot-scraper javascript /tmp/output.html \
'document.body.innerHTML.slice(0, 100)'
- try that
We don't need automation for that, we "achieve" that through our processes already. Specifically, software creation processes of large teams with many changing developers over long periods. Example (but they are not the only one): https://news.ycombinator.com/item?id=18442941 -- changing or adding anything becomes increasingly burdensome.
I would like to post that every time somebody warns of the dangers of AI for maintainability. We are long past that point, long before AI. Businesses made the conscious decision that it is okay for quality to deteriorate, they'll squeeze profits from it for as long as possible and then they assume something new has already come along anyway. The few business still relying in that technical-debt-heavy product are still offered service, for large fees.
AI is just more of the same. When it becomes too hard to maintain they'll just create a new software product. Pretty much like other things in the material world work too, e.g. housing, or gadgets, or fashion. AI actually supports this even more, if new software can be created faster than old code can be maintained that's quite alright for the money-making oriented people. It is harder to sell maintenance than something new at least once every decade anyway.
heh, Im using this font in my game. Picking fonts is hard, and I feel like I've just dipped my toe in the water so far. Im not 100% satisfied with the non-monospace font I use (Adobe Source Sans), but I have more important things to focus on right now
reply