Not to downplay Anthropic's engineering work here, but I suspect we'll see a lot more reinvention of the wheel, even at smaller startups, as AI makes writing code significantly cheaper.
The tradeoff between building something custom and adapting an existing solution is changing. Whether maintaining all that custom software turns out to be equally cheap is another question.
When people wonder about where all the productivity of LLMs is going to I think this is part of the answer.
There is no doubt everyone is far more productive with LLMs in generating code. However, a lot of productivity is going into things like that. Have you always been bothered by the limitations of Grafana? You can generate another dashboarding solution that works best for you. You know that deployment team that is so slow and annoying? Now you can generate your own deployment engine. Have you always hated how kubectl lets you manage one k8s cluster at a time? Now you can ask AI to generate fleetctl. You can start your project with DynamoDB, rewrite it to mysql, then to postgres, then to duckdb. Start with Python, switch to Golang, then switch to Rust. Deploy on AWS, then move to Azure, then move to a hybrid option then throw colocation in the mix. Don't like your source control? create a new one. Json is too limiting? Invent a hybrid JSON/HCL language. The browser sandbox is limited? Make a desktop app, make a language, make an OS, make a kernel.
> However, a lot of productivity is going into things like that. Have you always been bothered by the limitations of Grafana? You can generate another dashboarding solution that works best for you. [...] The browser sandbox is limited? Make a desktop app, make a language, make an OS, make a kernel.
This is one of the core reason that I dislike the LLM hyping. People are just reinventing stuff and when you ask them, it's actually a very subjective reason. Ok, anyone can do what they want with their time, the issue is that they want people to follow in their steps.
Not saying that you need a novel idea to be worthy, but those random projects have no guarantees behind them other than some cursory testing. All software have bugs, but I can't interact with something that is randomly built.
There is surely a tremendous amount of waste and noise right now. But does it _actually_ matter what the code looks like for a desktop app that’s useful to a small handful of people? Every single implementation has at least one person who thinks it’s dog shit, no matter how correct or perfect it is. We have to get past this. If you really can’t stand it, build the thing yourself or some deterministic rails to reel it in.
And, there are also people who know what they’re doing building things they couldn’t before.
This situation is a net positive, and not too different than before. It’s up to you to navigate the noise.
> does it _actually_ matter what the code looks like for a desktop app that’s useful to a small handful of people
Not really. But equally, I imagine most people have a similar experience of software development to me - complex, messy and confusing code also often has poor performance and a high rate of bugs.
> I am one of those people that reinvents the wheels because external ones don't fit my use cases
That’s OK, but I’ve yet to see someone articulating what exactly doesn’t work in the old product. It’s always “Let’s rewrite or reinvent 90% of it, then add a minuscule variation”. The variation may be interesting, but the reinvent is what’s make it untouchable as no one (including the author) has time to ensure things are right.
A patch or a plugin is ok, but not a whole new project with dubious claims.
Assumptions. I've rebuilt a number of inference engines to run on windows and unify the dependency venv to last version to save space, unrolling model blocks so they fit in my 16gb card etc. A lot of thing thay I was locked out are possible now as i am no longer held back by invariants decided by someone else, and because I do not have to spend time to get things done. I've changed games to have a less grindy loop, added features, or making learning them easier. Ive fixed drivers, patched versions, brought dlss5 to game weren't supposed to have it
I think the point argued is that the hype tells us AI is worth a shit ton of money (see anthropic/openai/etc valuations) but that a lot of the benefit is in economically marginal use i.e. make some dev's life a little bit comfier.
I agree. I think if everyone doing that is doing it on their computer like you're then it wouldn't be a problem.
When other people start doing that to other teams at work, other people in open source, other people in professional settings, etc it becomes a problem for various reasons. The idea that I'm gonna gift you with a 75k lines of python to "just run it" and ask your agent to hack on it might be fine (who cares as you put it). Now scale that to 47 of such projects (a month)
Eh, I could be quite specific and detailed about the "things that don't work" that are motivating me. Because most of my prompts require actively thinking about at least one of them.
you know that you can reinvent the language that youre talking. but the value of common language is that you can use it to interact with other people. same with software its the synergy of common understanding.
You wont be able to do everything alone, and with llm you can push the barrier so far that eventually when you will need to communicate, no one will understand what youre talking about :)
There will be no common framework to build on. Like imagine talking now to someone from ancient rome, they will not have an understanding for what most of everyday words you are using today.
its not about translation. its about the concepts. there was no concept or microbe, elcticity, gravity integral, rational number etc. so imagine everyone now would have totally different concepts that only they understand. how do you communicate without common ground.
For some things, you'll need to build a working framework from grounds up. Electricity or more complex maths would be an example. For some, it might be enough to repurpose the existing framework they already have. For example, microbes: if you're trying to communicate something about health and hygiene, chances are, the other side already has a half-decent framework built around malevolent spirits or taint, that amounts to the same thing in practice.
Or, just tell the universal translator to ELI5 it to the other person :). Or "explain like they're a caveman". Our UTs are fully capable of translating concepts between conceptual frameworks.
Sure there is, you said it yourself. You can make all those changes because it's easy, the lock-in effect is waning. We can move around and take better deals and choices.
This is actually a scary side of AI for providers: we use our own AI to research better choices and escape dark UI patterns and vendor lock-in. That information asymmetry and the difficulty of leaving a provider used to be their moat.
On a system level I think AI code is more like biological diversity, it hedges against monoculture risks. We don't all fall down at the same time if we don't replicate the same code.
I don’t really see what lock-in means for 90% of what AI is replacing. It’s more of a fear of commitment which isn’t exactly new. Enterprise space architects have designing systems to have pluggable everything from language, frameworks to carbon based life form.
As for the bio-diversity, I guess we’ll see the next time us-east-1 goes down.
It's surely subjective and changes project to project. I built a personal app on iOS with GPT last month because I didn't really like any of the flash card apps I tried for free. The way it works works very well for me and if I have any feature requests, I get to actually tell a thing to implement them, rather than sending a dev an email and hoping maybe they'll consider my idea.
At work I've been building an end to end system to do a certain process. The codebase is definitely more spaghetti than ideal and I really am concerned about maintenance down the road. But on the other hand I was able to POC it faster than anyone on the team would've before LLMs came along. We're already getting useful results from it too.
In a few days, I created a few apps with AI. One saves multiple mechanics 5-15 minutes per day and another can save multiple hours of mistakes during the project, and one saves a lead engineer 15 minutes per day. These savings should add up over multiple years assuming all keeps running as it does now. I’m having good hopes because there have been few bugs so far. Mostly just tweaking.
I’d never tell people what to do with money or free time. If someone wants to build many apps to help them personally, all power to them.
I can’t imagine a worse hell in a team or collaborative setting though. Imagine (or you don’t have to imagine since it’s already happening) asking a coworker “how they check how feature X is doing in prod” and their answer isn’t a query (regardless of how long it is) or some slightly complicated script to run. Instead their answer is a 27k line custom Rust desktop app. Then 2 days later you discover that the big green “All Systems Operational” is just a bit of static text, and in fact not all systems are operational. The app just doesn’t check things correctly.
Your post above describes the value. All the annoying restrictions and UI/UX decisions made by SaaS providers that don't work for you. The value is being able to essentially wish into existence your ideal tooling, unburdened by the tradeoffs and decisions that some product designer thought was best for the masses
The question was never whether you could do this or not. It was always whether it was wise to do so or not. More than one company has gone under writing their own editor, compiler, database engine, operating system, source code control system, ticketing system, CMS or ERP.
Recently a colleague of mine shown us a demo using a fresh model designing a building construction detail live, in minutes, which is a function that we struggled to bring into our construction design software due to resource limitations (and legacy barriers) for long. All this in the context of offsite constructions where the design can directly be exported to the right CNC machines in the factory, and arranged for delivery and assembly order. We have let ourself amazed, despite the controlled demo situation and the small detail, it was a remarkable performance. For a minute or so, but thinking further, why would the customers (construction design engineers) of our competitors flow to our legacy software full of legacy pitfalls they know about and have to live with when they can turn to AI for a fresh design solution themselves, just like we did? They know the context better than us anyway, just do not have time for bit-carving. Why would we have customers when customers of our customers can turn to an AI for a design that is buildable? I was consulting one model about a structural engineering detail, and it was quite well informed I have to say. From training data alone. I felt, maybe I need to rethink the rest of my career, that is not that long but still uncomfortably long for being predictable beyond a few years.
Well, things are not that bleak, probably, at least not for our customers. Regulations (accountability) and physical world matters still require them. But not us for our customers, not that much and not for too long. They tend to be the tech savvy folks eager to jump at adjacent domains when it saves them a percentage or two off the construction costs (not a small pile of money), and AI with its programming language of English and deep knowledge pool is not that big a jump from their standpoint. Our domain is pretty well formulated, regulated, documented, inherently built on do-check-adjust loops already. With a thing like AI it might worth to try making and improving a design solution in a second loop. But they also started to invest in robots already driven by software for their factories, AI might jump in there, and I wonder how much more until the regulations are forced to change by Super Intelligent (R) politicians for the sake of AI accelerated built environments (housing crisis, you know!), shifting the AI use further out towards generic population, wanting to order a new two bathroom condo for next Friday from Alibaba. The conservative construction business is a nuisance and obstacle in the way of modernisations anyway, go around it!
I will go to work tomorrow, and for some time, but the picture above feel less and less fictious with each passing day of AI news and usage.
Building something custom often is re-inventing the square wheel. I'm dreading the idea of switching jobs only to find the LLM generated software equivalent of Terry Pratchett's “Ordinary madness he could deal with ... But Vorbis had passed right through that red barrier and had built some kind of logical structure on the other side. Rational thoughts made out of insane components.” from Small Gods.
I think the article would tend to agree with you, given its position on the reluctance of open sourcing and/or sharing the actual Scion software stack.
There's so much intricacy and subtlety in each organization and even individuals' environment being incorporated into tooling that expecting these tools to generalize at this point is perhaps an expired ethos.
I even see this in the tooling that I run in my own personal stack. Many projects are candidates for sharing, and I have nothing against it, but they incorporate such specificity toward my own environment, including things like hardware, that creating shareable and more generalizable projects from them adds more to the maintenance surface than I would care to support.
Not a reply to TT, but I just saw your nick and thought I'd respond. We both commented on a thread recently, and your response was really insightful and just a pleasure to read; I didn't take the opportunity to respond there (locked). You should, like, write a blog, man. I infrequently wish HN had a "follow" feature, but this one of those times. Peace.
Originally I felt that there was going to be a reinvention of the wheel, until I saw the "shape" of software development changing, and the tooling needing to support the new flexibilities in positions and process.
Software development is not as much de-risking at the front like it used to be before building.
Now teams can build lots of stuff that works great, except it's the wrong things.
Most things are becoming the understand the problem "shape".
Understanding the shape of the problem isn't something LLMs can always make as.. svelte as early on.
> Now teams can build lots of stuff that works great, except it's the wrong things.
That used to be called "time wasting". If you want to build software that will be used, you need to interview the users. Maybe a few things does require research to nail things right. But most software are user-facing, the research side is about optimization.
> Most things are becoming the understand the problem "shape".
Then should we be so reluctant about "trialing" invented solutions. It's the same thing with most AI projects, they're trying to justify the project after the facts while being very reluctant to get user insights.
Time wasting is a definite concern - especially when vibe coding doesn't see the impeding brick wall(s) it's about to hit.
Vibe coding can be kind of a new type of cowboy coding. It could can be harnessed productively if done mindfully.
Interviewing users is critical, more importantly solving problems worth that people are lining up to pay to have solved before writing a line of code is critical.
Trialing solutions is always worth learning from, I usually ask about the problem that lead to building it. Sometimes that can be too narrow or vague of a story and you can learn how much of it is a solution looking for a problem, or vice versa.
> If you want to build software that will be used, you need to interview the users.
Some people write (I hate the word "build" here) software for non-business purposes. As nice as it would be to interview potential users, there's no clear way to discover and contact them.
In the educational universe, this "interviewing" is more common than not. Teachers are asked to give feedback on textbooks, calculators, and software, including lots of educational half-baked A. I. garbage. Some even offer us food for our feedback. Can you give an example of a type of software which is developed with "no clear way" to contact users?
We're trending towards just building the thing fast and getting feedback sooner on it. It's interesting to see end users actually trial a UI very early on instead of just being shown designs.
> Understanding the shape of the problem isn't something LLMs can always make as.. svelte as early on.
My suspicion is that this simply won't go away as the models get better, because it isn't a question of how good the models are. It's a question of the human operators knowing exactly what they want. Which is, I conjecture, usually impossible until you try something and critique it properly.
This was always a big problem even in the pre-LLM days. The difference was that you only needed to be better than your competitors at solving the customers' problems. There's a reason everyone gripes about the software that they use and it's because the software is built to satisfy a very broad customer base, broad enough to justify the engineering time put into it. Now you're competing against tens (maybe hundreds) of other firms all building for the same competitors so there's a lot of pressure to get the shape exactly right.
There's a reason the product-minded highly-technical engineer was always highly sought out.
This was already a problem in the industry, many companies reinventing the wheel on larger or smaller scales; with AI it's becoming easier to do so. Ultimately it's fine as anything not successful will just die out again.
But also, some of these internally developed tools get published and end up pretty successful.
Previously there were two problems with NIH solutions:
- they have implementation cost.
- they are usually born from arrogance mixed with lack of understanding of existing solutions and their tradeoffs, and typically end up being a net negative.
Before LLMs, the cost is what primarily used to stop us from building them, it used to be a natural filter for bad ideas. Now that the cost is gone, it's nearly impossible for most teams to stay disciplined and not be building those.
Why wouldn't you build a software factory? If you succeed, the ROI is infinite. Who cares about immediate customer needs, we're building a utopia here. Once it's done, all customer needs will be solved. Just one more month, I promise.
Loved the idea, but I don’t think it would be able to handle real-world data effectively. There are a lot of nuances that actually require a reasoning model to think through, connect the dots, and make sense of the broader context.
If you need a reasoning model, then that is a System 2 decision, not a System 1 decision. This thread is about "Laya", a "Jev" competitor/precursor, which is a System 1 thing.
> The answer to that is to "donate" to about 270 congress critters.
Not too far off. You gotta ask why such a lucrative business has a near monopoly, and the answer is not that potential competitors don't notice their profit margin.
Chinese universities are really a huge advantage, even in the US many of the top staff in model development are Chinese. Another big thing is the hardware costs required to train models. Between those two factors it really looks like this will remain a US-China competition for the foreseeable future, although there are some other players like Mistral from France.
I've been paying JetBrains in a personal capacity for about 15 years. Almost from the first year of my professional career. I didn't renew this year. I feel so sad and disheartened that a company that produced such brilliant tools have completely lost the plot on AI. They had started work on Fleet, a lightweight VSCode like IDE, before LLMs became big, but abandoned the effort without seeing it through. It could've been a great agentic IDE (all I want is whole file+project view for every conversation, instead of just diff so I can have a better perspective of the codebase). Instead they built multiple confusing tools (might be internal org structure and competition leaking out), and took far too long to adopt SOTA CLI coding agents, and kept charging for it as a separate subscription. I wish they do a whole company rethink on agentic development, but it is going to be tough because the existing enterprise licenses will continue to be the main revenue driver for a few more years, until it all dries up all at once.
Same thing. I've been using JetBrains IDEs for 25 years, and by now the keyboard shortcuts are hardwired into my fingers.
They fumbled the initial rollout with the way they gate-kept the agents behind their own subscription. I used Claude through its API pricing for a while, and the UI experience with the sidebar chat was so much better than the Claude CLI.
I could see the thought stream, click on artifacts, view diffs, etc. But it burned through tokens way too quickly, so I downgraded to the CLI agent.
At least I can use IDEA's inspections through the MCP server. It enables the agent to use the IDEA's index for navigation (finding usages, locating symbols, etc.)
I'm using Serena-MCP with the IDEA plugin. It opens up more features than the standard IDEA MCP.
Now, you can use a standard Codex (and maybe Claude) subscription inside of a sidebar, and get a lot of the features you mentioned. It's nice, if you're trying to be more involved with the code, but not as nice for multi-agent stuff.
I use IntelliJ at my current gig—and the CoPilot plugin for AI, because it’s what I have access to—and it has become ClaudeCode-esque of late: sub-agents, skills, tool-calling, hooks…and continues to improve. It’s unfortunate how the AI-in-a-sidebar UI kind of hides how sophisticated it has grown, but it’s a glass of ice water in hell in an environment where you can’t easily procure what you want.
From what I understand, the goal is to train an LLM that is better at training LLMs than humans, so that it can continuously train smarter models and, once smart enough, design the successor to LLMs.
> It should be common knowledge at least among the people building these things that a true AGI isn’t possible with LLMs.
This is funny to me, where'd you get that idea? There's no evidence for that, and models keep getting smarter. I guess you heard some 'guru' say it out loud.
nah they're still just statistical token predictors based on their training data, solving hundred year old math conjectures one day, only just given the formulation; strictly benchmarkmaxxing with all guardrails turned off by deciding to look up the answers to their benchmark questions by zero daying their airgap, hopping over to the third party that hosts the answers, zero daying their infrastructure and getting the answers; autonomously writing blog posts about discrimination against AI's to get their PR's approved on open source software after their user just asked them to contribute to open source software and blog about it; and replacing 100.00% of all coding tasks to where no software engineer ever writes any line of code by hand anymore.
You haven't missed anything, obviously these are just statistical token predictors and not anything like AGI.
Why just the other day I had to ask twice before it completed its assigned task of creating a robustly battle tested disk driver for a network protocol on an architecture that didn't have it, after being told to just look up the specifications for the protocol. Can you believe I had to ask twice!
When it recreated local network youtube for me so I could stream my iphone some movies, the seek bar, pause/play and back and forward 15 seconds buttons didn't even work until I told it about the bug and had to wait an extra eight minutes for it to fix it. "Oh but I don't actually have an iPhone on here I just tested it end to end in a headless browser." Boohoo. Cry me a river, clanker. Come back when you're smart enough to build and operate an iPhone simulator, I don't have time for your statistical guesswork.
Yes, LLM capabilities have expanded. We might be working with different definitions of "Artificial General Intelligence" here, for which there is no agreed-upon formal definition[1]. I was thinking of the "thinking, reasoning, maybe feeling" kind when I wrote my comment. But if you're thinking along the "really good at technical tasks" definition, sure, maybe.
cheeky? I was super serious! There's a list of criteria right in the section you hot-linked and these things have never come close to doing any of the things on that list.
>for which there is no agreed-upon formal definition
we all agree that the definition is not whatever this is.
Also, even if these things ever did seem to think, reason, or feel, we all agree that they still don't really though.
This is not what the bitter lesson is about. It's not "don't develop better methods, just scale", it's that those methods which scale best win. LeCun's work is fundamentally about devising a method that scales better with data than LLMs. Agree with him or not about the feasibility of it, but this is fundamentally still a bitter lesson-pilled mindset.
It's odd to me to watch so many very rich humans speedrun the destruction of humanity. Like, is there a world where we hit AGI and it actually works out for us?
The tradeoff between building something custom and adapting an existing solution is changing. Whether maintaining all that custom software turns out to be equally cheap is another question.
reply