Just yesterday I tried to feed it a simple HTML page to extract a selector, I tried it with GPT-4-turbo, I tried it with Claude, I tried it with Groq, I tried it with a local LLama2 model with 128k context window. None of them worked. This is a task that while annoying, I do in about 10 seconds.

Sure, I'm open to the possibility that in the next 2 - 3 days up to a couple of years, I'll no longer do manual coding. But honestly. After so much hype, I'm starting to grow a bit irritated with the hype.

Just give me a product that works as advertised and I'll throw money your way because I have a lot more ideas than I have code throughoutput!

The problem with this entire space is that we have VC hype for work that should ultimately still be being done in research labs.

Nearly all LLM results are completely mind blowing from a research perspective but still a long way from production ready for all but a small subset of problems.

The frustrating thing, as someone working in this space awhile, is that VCs want to see game changing products ship overnight. Teams working on the product facing end of these things are all being pushed insanely hard to ship. Most of those teams are solving problems never solved before, but given deadlines as though they are shipping CRUD web apps. The kicker is that despite many teams doing all of this, because the technology still isn't there, they still disappoint "leadership". I've personally seen teams working nights and weekends, implementing solutions to never before seen problems in a few weeks, and still getting a thumbs down when they cross the finish line.

To really solve novel problems with LLMs will take a large amount of research, experimentation and prototyping of ideas, but people funding this hype have no patience for that. I fear we'll get hit by a major AI winter when investors get bored, but we'll end up leaving a lot of value on the table simply because there wasn't enough focus and patience on making these incredible tools work.

Yeah, I remember speech recognition taking decades to improve, and being more of a novelty - not useful at all - even when it was at 95% accuracy (1 word in 20 wrong). It really had to get almost perfect until it was a time saver.

As far as coding goes, it'd be faster to write it yourself and get it right first time rather than have an LLM write it where you can't trust it and still have to check it yourself.

It may well turn out that LLMs are NOT the path to AGI. You can make them bigger and better, and address some of their shortcomings with various tweaks, but it seems that AGI requires online/continual learning which may prove impossible to retrofit onto a pre-trained transformer. Gradient descent may be the wrong tool for incremental learning.

That being said, something is definitely coming. 50% correctness would probably be well worth using — simple copy/paste between my editor and GPT4 has been useful for me, and that's much less likely to completely solve an issue in one shot — and not only will small startups doing finetunes be grinding towards better results... The big labs will be too, and releasing improved foundation models that the startups can then continue finetuning. I don't think a new AI winter is on the horizon yet; Meta has plenty of reason to keep pushing out better stuff, both from a product perspective (glasses) and from an efficiency perspective (internal codegen), and OpenAI doesn't seem particularly at risk of stopping since Microsoft is using them both to batter Google on search (by having more people use ChatGPT for general question answering than using Google search), and to claw marketshare from Amazon in their cloud offerings. Similarly, some AI products have already found product/market fit; Midjourney bootstrapped from 0 to $200MM ARR (!) for example, purely on the basis of monthly subscriptions, by disrupting the stock image industry pretty convincingly.

I also have two crypto-bro friends that are hyping it up without having anything to show for it. Which is why I'm sort of complaining about they hype surrounding it. I agree with your post to a large extent. This is not production ready technology. Maybe tomorrow.

These generally don't require advanced logic or thought, though they can require some moderate reasoning ability to summarize two slightly conflicting text extracts.

Lots of corporate work would be enhanced by better summarization, better information dissemination, and better text extraction. Most of it is pretty boring work, but there's a lot of it.

VC hypes seem to want to mostly focus on fantastical problems, though, which sound impressive at dinner parties but don't actually work well.

If you're a VC, do you want to talk about your investment in a company that finds discrepancies in invoices, or one that self-writes consumer iPhone apps?

Only one of those is actually doable today.

...this is what happened in 99-2000. It took 3-7 years for the survivors to start making it usable and letting the general public adjust to a new user paradigm (online vs on PC).

This is an important lesson that all SWEs should take to heart. Nobody cares about your novel algorithm. Nobody cares about your high availability architecture. Nobody cares about your millisecond network latency optimizations. The only thing that anyone actually using your software cares about is "Does the screen with lights and colors make the right lights and colors that solve my problem when I click on it?". Anything short of that is yak shaving if your role is not pure academic R&D.

Our jobs are safe. I would even expect more "beginners" to try something with AI and then need an actual programmer to help them

( At least, if they are unwilling to invest the time in development and debugging themselves

Ps. Probably all the given examples are in top 3 most popular programming languages.

Amateurs at Python, Fortran, C, C++ and all programming languages. Amateurs at car engineering, airplane engineering, submarine engineering etc. Amateurs at human biology, animal biology, insect biology and so on.

I don't know anyone who is an amateur at everything.

In submarine engineering, they will be able to design and construct it in some way like 3d print it, and the submarine will be able to move into the water for some time before it sinks. Yeah, probably for submarines a higher percent should be achieved before they are really useful.

Honestly, LLMs can't even get language right. They produce generic, amateurish copy that reads like it's written by committee. GPT can't perform to the level of a middle market copywriter or content marketer. I am convinced that people who think LLMs can write have simply not understood what professional writers do.

For me the "plateau of productivity" after the disillusionment has been using LLMs a bit like search engines. Quick standalone summaries, snippets or thoughts. A nice day-to-day productivity boost, but nothing that's going to allow me to work less hard.

I've had the same experience as well. I heard tons of people clamoring about the ability for LLM's to write SEO copy for you and how you can churn out web content so much faster now. I tried using it to churn out some very specific blog posts for an aborist client of mine.

The results were really bad. I had to re-write and clarify a lot of what it spit out. The grammar was not very good and it was really hard to read with very poorly structured sentences that would end aburptly and other glaring issues.

I did this right after a guy I play hockey with said he uses it all the time to write emails for him and pays the monthly subscription in order to have it write all kinds of things for me every day. After my trial, I was really wondering how obvious it was that he was doing that and how his clients thought about him knowing how poorly the stuff these LLM's were putting out.

>After my trial, I was really wondering how obvious it was that he was doing that and how his clients thought about him knowing how poorly the stuff these LLM's were putting out.

I feel the same way about this stuff as when devs say they push out LLM code with no refactoring or review. Ah, good luck!

GPT's rigid "robot butler" style is not "just how LLMs write". OpenAI deliberately tuned it to sound that way. Even much weaker models that aren't tuned to write in a particular way can easily pass for human writing.

It's funny to see, because developers are so sensitive about being treated like code monkeys by their non-technical colleagues. But these same devs turn around to treat other professionals as word monkeys, or pixel monkeys, or whatever else. Not realizing that they are only seeing the tip of the iceberg of someone else's profession.

Professional writers don't take prompts and shit out words. They work closely with their clients to understand the important outcomes, then work strategically towards them. The dead giveaway of LLM writing isn't the style. It's the lack of coherent intent behind the words, and low information density of the text. A professional writer works to communicate a lot with very little. LLMs work in the opposite way: you give it a prompt, then it blows it out into verbiage.

Sit down for coffee with a professional copywriter (not the SEO content marketing spammers), and see what they have to say about LLMs.

Personally, I group all these things under 'style'. Perhaps, i should have used, 'presentation' instead. You've latched on that specific word and gone off. Point is that the post-training of these models, especially GPT from Open ai is doing a lot to how the writing (the default at least) presents long strings of text. Like how GPT-4 is almost compelled to end bouts of fiction prematurely in sunshine and rainbows. That technically isn't style but is part of what i was talking about.

>A professional writer works to communicate a lot with very little. LLMs work in the opposite way: you give it a prompt, then it blows it out into verbiage.

There's no reason you have to work this way with an LLM.

No, I haven't. I'm not talking about style, but something deeper. What I'm talking about is something you don't even seem to realize exists in professional writing - which is why you keep thinking I'm misunderstanding you when I am not.

I've worked with professional writers, and nothing in the LLM space even comes close to them. It's not a matter of low quality vs high quality, or benchmarking, or style. It's simply an apples and oranges comparison.

The economics of LLMs for shortform copy will never make sense, because producing the words is the cheapest part of that process. They might become the best way for writers themselves to produce longform copy on the execution side, but they can't replace the writer's ability to work with the client to figure out exactly what they are trying to write, and why, and what a good result even looks like. And no, this isn't a prompting issue, or a UI issue, or a context window length issue, or anything like that.

Elsewhere in this thread someone mentioned how invaluable LLMs are for producing internal business copy. I could easily see these amateur writing tasks being replaced by LLMs. But the implication there isn't that LLMs are any good at writing, but that these tasks don't require good writing to begin with.

Then the humans can make a heroic return, write surgical prose like Hemingway to slice through the AI drivel, and keep collecting their paychecks.

Bonus points if you can translate this analogy to software development...

In some circles, that's actually a wonderful achievement. :p

And it only took one of the most computationally expensive processes ever devised by man.

If you use GPT-4 day-to-day, you've probably encountered this sense of a capability wall before. The point where additional prompting, tweaking, re-prompting simply doesn't seem to be yielding better results on the task, or it feels like the issue is just being shifted around. Over time, you develop a bit of a mental map of where the strengths and weaknesses are, and factor that into your workflows. That's what writing with LLMs feels like, compared to working with a professional writer.

Most writers have already realized that LLMs can't write in any meaningful way.

[1]https://github.com/xenodium/chatgpt-shell/blob/bf2d12ed2ed60...

Or just write the damn thing yourself.

Ask it to "rewrite that in the tone of an English professor" or "rewrite that in the style of a redneck rapper" or "make that sound less like generic ad copy". Get into an argument back and forth with the LLM and tell it the previous response is crap because of XYZ.

The model names, version numbers or who makes them are irrelevant.

1) copilot is a terrific auto complete, and writes tremendous amounts of repetitive boilerplate

2) copilot can help me kickstart writing some complex functions starting from a comment where I tell it what is the input and expected output. Is the implementation always perfect or bug free? No. But in general I just need to review and check rather than come up with the instruction entirely.

3) copilot chat helps me a lot in those situations where I would've googled to find how to do this or that and spent a lot of time with irrelevant or outdated search results

4) I have found use cases for LLMs in production. I had lots of unformatted plain text that I wanted to transform in markdown. All I needed to do is to provide few examples and it did everything on its own. No need to implement complex parsers, but make a query to OpenAI with the prompt and context. Few euros per month in OpenAI credits is still insanely cheaper than paying tons of money in writing and maintaining software by humans for that use case.

5) It helps me tremendously when trying to learn new programming languages or remembering some APIs. Writing CSS selectors is actually a very good example. But I don't feed it an entire HTML as you do, I literally tell him "how do I target the odd numbered list elements that are descendants of .foo-bar for this specific media query". Not sure why would you need to feed it an entire HTML.

6) LLMs have been extremely useful to generate images and icons for an entire frontend application I wrote

7) I instruct him to write and think about test cases about my code. And it does and writes the code and tests. Often thinks about test cases I would've never thought of and catches nice bugs.

I really don't buy nor think it can write too much on its own.

The promise of it writing anything but simple boilerplate, I find it ridiculous because there's way too much nuance in our products, business, devices, systems that you need to follow and work on.

But as a helper? It's terrific.

I'm 100% sure that people not using these tools are effectively limiting themselves and their productivity.

It's like arguing you're better off writing code without a type checker or without intellisense.

Sure you can do it, but you're gonna be less effective.

1) Autocomplete is more often than not what I want or pretty darn close.

2) Sometimes I need a discrete function that I am not sure how I want to write. I use a prompt with 3.5/4 inside of my IDE to ask it to write that function.

It is definitely not writing complete programs any time soon but I can see where it's heading in the near term. Couple it with something like RAG to answer questions on library/api implementations. Maybe give it a stronger opinion about what good Python code looks like.

For the naysayers I don't know how you use it but it is certainly useful enough for me to pay for.

Especially #5. I'm certain that I've been at least 10x more productive in learning new tools since chatgpt hit the scene. And then knowing it helps so much with that has had even more leverage in opening up possibilities for thing I'm newly confident in learning / figuring out in a reasonable amount of time. It is much easier for me to say "yep, no big deal, I'm on it" when people are looking for someone to take on some ambiguous project using some toolset that nobody at the company is strong with. It solves the "blank page" issue with figuring out how to use unfamiliar-to-me-but-widely-used tools, and that is like a superpower, truly.

It's pretty decent for "happy path" test cases, but not that good at thinking of interesting edge or corner cases IME, which comprise the most useful tests at least at the unit level.

I'm pretty skeptical of #4. I would be way too fearful that it is doing that plain text to markdown transform wrong in important-but-non-obvious cases. But it depends on which quadrant you need with respect to Type I vs. Type II errors. I just never seem to be in the right quadrant to rely on this in my production projects.

The "really good intellisense" use cases #1-#3 also make up a "background radiation" of usefulness for me, but would not be nearly worth all the hype this stuff is getting if that were all it is good for.

I agree. I have it active on VSCode and enjoy it. It has introduced subtle bugs but the souped up autocomplete is nice.

> 2) copilot can help me kickstart writing some complex functions starting from a comment where I tell it what is the input and expected output. Is the implementation always perfect or bug free? No. But in general I just need to review and check rather than come up with the instruction entirely.

I don't find it very useful for anything non trivial. If anything I found it more useful for generating milestones and tasks for a product, than even making a moderately complex input -> output without me having to check it in a way that annoys me.

> 3) copilot chat helps me a lot in those situations where I would've googled to find how to do this or that and spent a lot of time with irrelevant or outdated search results

I find I don't use copilot chat, almost at all. Nowadays I prefer to go to Gemini and throw in my question.

> 4) I have found use cases for LLMs in production. I had lots of unformatted plain text that I wanted to transform in markdown. All I needed to do is to provide few examples and it did everything on its own. No need to implement complex parsers, but make a query to OpenAI with the prompt and context. Few euros per month in OpenAI credits is still insanely cheaper than paying tons of money in writing and maintaining software by humans for that use case.

This is mostly what I'm using it for in this current project. It does it job nicely but it's very far away from replacing myself as a programmer. It's more like a `fn:magic(text) -> nicer text`. This is a good use case. But it's a tool, not a replacement.

> 5) It helps me tremendously when trying to learn new programming languages or remembering some APIs. Writing CSS selectors is actually a very good example. But I don't feed it an entire HTML as you do, I literally tell him "how do I target the odd numbered list elements that are descendants of .foo-bar for this specific media query". Not sure why would you need to feed it an entire HTML.

Because I get random websites with complex markup, and more often than not every page has its unique structure. I can't just say give me `.foo-bar` because `.foo-bar` might not exist. Which is where the manual process comes in. Currently, I'm using hand crafted queries that get fed into GPT / Claude / LLama, but the actual query is what I wanted it to do.

> 6) LLMs have been extremely useful to generate images and icons for an entire frontend application I wrote

I'm very curious how this behaves in different resolutions. There's a reason vector graphics are a thing. I've used it for this purpose before but it doesn't compare to vectorial formats.

> 7) I instruct him to write and think about test cases about my code. And it does and writes the code and tests. Often thinks about test cases I would've never thought of and catches nice bugs.

What is the context size of your code? It works for trivial snippets but as soon as the system is a bit more complex, I find that it becomes irellevant fairly fast.

> The promise of it writing anything but simple boilerplate, I find it ridiculous because there's way too much nuance in our products, business, devices, systems that you need to follow and work on.

> But as a helper? It's terrific.

> I'm 100% sure that people not using these tools are effectively limiting themselves and their productivity.

Totally agree. But I'm not complaining about its usefulness. I'm a paying user of LLM systems. I use them almost every day. They're part of my products. But this particular hype about it replacing ... me. I don't buy. Yet. It could come tomorrow and I'd be happier for it.

Rather, your “problem” is that you’re likely not writing uninteresting cookie cutter boilerplate that everyone can do and has done hundreds of times. The current crop of AI is cool for coding demos, not for solving real relevant problems.

> Just give me a product that works as advertised and I'll throw money your way

The people hyping this crap only care about the second part of that sentence. The first one is an afterthought.

- replacing StackOverflow and library documentation

- library search

- converting between formats and languages

- explaining existing code/queries

- deobfuscating code

- explaining concepts (kinda hit or miss)

- helping you get unstuck when debugging or looking for solution (‘give me possible reasons for …’)

I feel like many of this things require asking the right questions, which assumes certain level of experience. But once you reach this level, it’s an extremely valuable assistant.

I find it horrible at replacing library documentation

> I feel like many of this things require asking the right questions, which assumes certain level of experience. But once you reach this level, it’s an extremely valuable assistant.

I've been using LLM products since incipience. I use them in my daily work life. It's a bit tiring hearing this 'right questions', 'level of experience' and 'reach this level'. Can you share anything concrete that you achieved with ChatGPT that would blow my mind?

I keep hearing this 'you need to ask the right kind of questions bro' from people that never build a single product in their life, and it makes me question my ability to interact with LLM but I never see anything concrete.

GPT reconstruction:

The dream reaches its peak when you meet the "final boss" of the game: an entity that embodies the ultimate barrier to communication. To overcome this obstacle, you must synthesize everything you've learned about "it" in the dream and present a coherent vision that is true to yourself. As you articulate your final understanding of "it", the maze dissolves around you, leaving you in front of a giant mirror. In this mirror, you see not just your reflection but also all the characters, passions, and ideas you encountered in the dream. You realize that "it" is actually a reflection of yourself and your ability to understand and share your inner world. The dream ends with the title revealed, "Game to Learn to Communicate about It and Its Player", meaning the whole process was a metaphor for learning to know and communicate your own "it" - your personality, thoughts, and emotions - with others, and that you are both the creator and the discoverer of your own communication game.

My note:

The continuation of the dream corresponds to an abrupt change of scene. I find myself in my bed, in the dim light of my room, facing a mysterious silhouette. As I repeatedly inquire about its identity, I stretch my hands towards its face to feel its features as I cannot clearly see them. Then, a struggle begins, during which I panic, giving the dream a nightmarish turn. Noticing that the dark figure mirrors my movements, I realize it's myself. Suddenly under my duvet and as I struggle to get out, I feel jaws and teeth against the sheets. I call out for my mother, whom I seem to hear downstairs, and that's when my vision fades, and I see the dream's source code displayed behind. It consists of ChatGPT prompts shared on the lime green background of an image-board. At the bottom, I then see the dream's title: "Game to learn how to communicate about It and its player."

Library documentation? Yeah, it doesn't really save time when GPT makes up functions and libraries, making me check the docs anyways...

I was initially hopeful but I find it gets in my way for anything not trivial.

That behavior is now vanishingly-rare, at least in GPT4.

Did you use an AI assistant while making that list?

I think people want it to completely replace developers so they can treat programming as a magic box, but it will probably mostly help big picture architecture devs compete with people who are really good at Leetcode type algorithm stuff.

I am going to work on something after this pot of coffee brews that I simply could not produce without chatGPT4. The ideas will be mine but the most of the code will be from chatGPT.

What is obvious is different skill sets are helped more than others with the addition of these tools.

I would even say it is all there in the language we use. If we are passing out "artificial intelligence" to people, the people who already have quite a bit of intelligence will be helped far less than those lacking in intelligence. Then combine that with the asymmetry of domains this artificial intelligence will help in.

It should be no surprise we see hugely varied opinions on its usefulness.

The competition should be happening in the other direction.

In the realm of coding here are a few things its really good at:

- Translating code, generating cross language clients. I'll feed it a golang single file API backend and tell it to generate the typescript client for that. You can add hints like e.g "use fetch", "allow each request method to have a header override", "keep it typesafe, use zod", etc

- Basic validation testing. It's pretty good at generating scaffold tests that do basic validation (Opus is good at writing trickier tests) as your building.

- Small module completion. I write an interface of a class/struct with it's methods and some comments and tell it to fill in. A recent one I did looked something like (abbreviated):

type CacheDir struct { dir string, maxObjectLifetime: Duration, fileLocks sync.Map }

type (cd *CacheDir) Get(...)

type (cd *CacheDir Set(...)

type (cd *CacheDir) startCleanLoop()

Opus does a really good job generating the code and basic validation tests for this.

One general tip: you have to be comfortable spending 5 minutes crafting a detailed query assuming the task takes longer than that. Which can be weird at first if you take yourself seriously as a human.

Note that I hadn't been able to do much of this with GPT-4 Turbo with with Claude Opus it really feels capable.

I'm talking about 4-turbo, 4-turbo preview and self hosted LLama2. What in God's name is not SOTA about this?

So what is the SOTA, in your opinion?

It's that hype-churn marketing that is a motivating factor for the groups to innovate, much like Formula 1. It might be distasteful, but that doesn't mean it isn't working.

Good fucking Christ, Opus has been out for like 8 days, I've had holidays longer than that!

Are they considerably better than existing non-AI tools + manual coding for this? In VSCode and Visual Studio, when working with an interface in C# for example, I can click two context menus to have it generate an implementation with constructors, getters, & setters included, leaving only the business logic code to write manually. You've mention you have to describe to the AI in comments, and then I assume spend time on a step to verify the AI has correctly interpreted your request & implemented.

I can definitely see the advantage for LLMs when writing unit tests on existing code, but short of very limited situations, I'm really finding it difficult to find the 55% efficiency improvements claimed by the likes of GitHub's AI Copilot.

Totally agree on the 55% figure being hogwash.

Can any convert a native iPhone app to an Android one?

Like, by now you'd think it would be inevitable that we wouldn't be writing software in a text-editor or IDE. Everything else we do on a computer is more graphical rather than textual, with the exception of software development. Why is that?

Part of the reason why I'm kind of bearish on AI is because it seems like we could have replaced written code with GUI diagrams as far back as the 80s, or at the very least in the early 2000s, and it seems like something that should have obviously caught on given that would probably be much easier for the average person. Again though, curiously, we're still using text editors. Perhaps despite the popularization of AI no-code builders we'll still see that the old model of hiring someone good at writing code in a text-editor remains largely unchanged.

Makes me wonder if there's just something about the process that we overlook, and if this same something could frustrate attempts at automating the process of writing code using AIs as much as it frustrated our attempts at capturing code using graphical symbols.

I don't think most people are building landing pages anymore by handwriting code anymore. Same with blogs (eg. Wordpress). There are MVPs of successful businesses that've been built by Bubble.io. Internal dashboards and such can definitely be built without code such as via Retool or Looker or whatever.

WYSIWYG obviously makes sense for frontend, but less so for backend. For backend code I don't really see how some visual drag and drop editor could make for a better interface than code. And even if it could, the advantage of code is that it's fully customizable (whereas with a GUI you're limited by the GUI), and text itself as a medium is uniform and portable (eg. easy to copy and paste anywhere).

Not to say that we can't create better interfaces than text, but I do think some sort of augmentation on top of a code editor is probably a more realistic short-term evolution, similar to VSCode plugins.

But the “you won’t need to code” reminds me “you won’t need to learn to drive”.

It’s the messy interface with the real world in both cases that basically requires AGI.

If AGI is just a decade off then, yep, I won’t need to code. But a decade is a long time and, more importantly, we’re probably more than a decade away.

And even if it is “just round the corner”, worrying about not needing to code would be worrying about deckchairs on the titanic. AGI will probably mean the end of capitalism as we know it, so all bets are off at that point.

It’s wise to hedge a little but also realise that to date AI is just a coding productivity boost. The size of the boost depends on how trivial the code is. Most of the code I write isn’t trivial and AI is fairly useless at that, certainly it’s faster and more accurate to write it myself. You can get a 50% boost if you’re writing boiler plate all day, but then you have to wonder why you’re doing that in the first place.

The point at which someone like myself is allowed to become aware that a company has developed that level of AI is well beyond the point of no return.

https://pavpanchekha.com/blog/chatgpt-herbie.html

> I wouldn't use it—sanity-checking its algebra is a lot of work, but even if you fixed that up, the high-level ideas typically aren't that good either.

This has been exactly my experience with chatgpt as well.

I highly doubt you're the dumb one here.

Write me the molecular simulation boilerplate because these crappy tools all have their own esoteric DSLs, then I tweak the parameters to my use case, avoiding the busywork -

e.g. "Write me a simulation for methane burning in air"

Gives me a boilerplate, I modify the initial conditions (concentrations, temperatures, etc) and then deploy. Have the LLM do the busy-work, so I dont have to spend ages reading docs or finding examples just to get started.

Now deploy to a stable environment. Thats what I'm trying to help with by building https://atomictessellator.com

And I'm sure the reason for that is the garbage input. From time to time I have to perform quantitative code analyses in our so called enterprise repositories. And the results are shocking every time. I have found an extremely poor SQL code block to type cast many columns in hundreds of projects. It was simply copied again and again even though the casting was no longer necessary.

The training base should be sufficiently qualified (and StackOverflow ranking is obviously not enough).

But unfortunately it's probably too late for that now. Now inexperienced programmers are undercutting themselves with poor AI output as training input for the next generation of models.

I hack on them till I get something sort of satisfying.

> You'd be surprised how much better your output is using prompting techniques tailored for your goal.

The biggest problem I encounter is context length, not necessarily the output for small inputs. It starts forgetting very fast, whether it's Claude, GPT+ or other self hosted models I've tried.

Also, current crop of LLMs are not there yet for large/largish projects. GPT4 is too slow and expensive, while Groq is superfast but open source models are not quite there yet. Claude is somewhere in the middle. I expect somewhere in the next 12 months there's going to be a tipping point where they will be capable, fast, and reliable enough to be in wide use for coding in this style[1].

[0] I have an AI horse in the game with http://pythagora.ai, so yeah I'm biased [1] It already works well for snippet-level cases (eg GitHub copilot or Cursor.sh) where you still have creative control as a human. It's exponentially harder to have the AI be (mostly) in control.

But roughly on par with what could be expected of today's junior software developer (unaided by AI)? Definitely.

Almost no products fit this description, and if they do then the marketing department is getting fired.

Does a Mcdonalds burger look like the picture?

If you go in with a healthy dose of cynicism IMO LLMs can impress. I’d call it a better google search and autocomplete on steroids.

Sometimes? But I don't go to McDonalds for the loss function between the picture and actual product. I go for the fast food and good taste (YMMV).

> If you go in with a healthy dose of cynicism IMO LLMs can impress.

I use them everyday in one way or another. But they're not replacing me coding today. Maybe tomorrow. And I go in with a healthy dose of optimism when I say this.

> I’d call it a better google search and autocomplete on steroids.

Sure, but this particular discussion is not about its ability to replace Google Search and / or Autocomplete.

It actually does in the countries that require it. You know you can write ACTUAL "truth in advertising" laws right?

Yeah, AI will do the easy and fun jobs for you. You will only need to care difficult decisions that you're going to be responsible for. What a wonderful world...

This site has honestly been absolutely useless when discussing new technology now. No excitement, no curiosity. Just constantly crapping on anything new and lamenting that a brand new technology is not 100% perfect within a year of launch.

Remove "Hacker" from this site's name, because I see none of that spirit here anymore!

The trouble is, there's still nowhere better.

I'm quite wary of the long-term implications and downstream effects of that occurring at scale. AI is typically presented as "the human's hands are still on the wheel," but in reality I think we're handing the wheel over to the AI -- after all, what else would the endgame be? By definition, the more it can do without requiring human intervention, the "better" it is. Even if replacing people isn't the intention, I fail to see how any other effect could usurp that.

Assuming AI keeps developing as it has been, where will we be in 20 years? 50? Will anyone actually have the knowledge to evaluate the code it produces? Will it even matter?

Perhaps it's because Dune is in the air, but I'm really feeling the whole "in a time of increased technology, human capabilities matter more than ever" thing it portrays.

TBH I'm more surprised when I don't need to help it now. After about 3 times where it cycles between incorrect attempts I just do the job myself.

I disabled copilot since it consistently breaks my flow.

What is worse is that seems to be leading to a self-amplifying feedback loop, where people not up to speed enough with the models try to use them, fail and give up making them fall even further behind.

For every technological advancement, artisans are the first to be made obsolete.

Sure we have landfills full of unworn textiles, the market says its good, but overall, we keep destroying what allows humans to seek meaning.

Our governments and society have made it clear, if you don't produce value, you don't deserve dignity.

We have outsourced art to computers, so people who don't understand art can have it at their fingertips.

Now we're outsourcing engineering so those who don't understand it can have it done for cheap.

We hear stories of those who don't understand therapy suggesting AI can be a therapist, of those who don't understand medicine suggesting AI can replace a doctor.

What will be left? Where will we be? Just stranded without dignity or purpose, left to rot when we no longer produce value.

I ask this question often, with multiple contexts, but to what end? Who benefits from these advancements? The CEO and shareholders, sure, but just because something can be found for cheaper, doesn't mean it improves lives. Our clothes barely last a year, our shoes fall apart. Our devices come with pre-destined expiration dates.

Where will we be in the future? Those born into money can continue passing it around, a cargo cult for the numbers going up. But what about everyone else?

I'm not saying that things were sunshine and rainbows pre-industrialization, but there's some level of analysis to be done on the durability and value of a handcrafted piece of clothing, the care that goes into maintaining it, the value of a local economy, and the other side where you're forced to buy cheap items that degrade at a far faster rate.

If a town's local businesses are put out by a new walmart's ability to carry low prices, does the town truly come out ahead with those low prices? Or does Walmart simply extract more money from the town than it returns, leaving the town worse off?

Labor has never been to afford so much luxury as the modern day.

Consider how you can't really get by in most of America without a car because we designed our cities to require them. It would be a mistake to conclude that, because life is harder in a car-optimized society without a car, society must be better off optimizing for cars.

Then the majority of us spent our time and energy in factories. That got automated during the Industrial Revolution. So we moved to service and knowledge jobs.

And now we're potentialy seeing the start of automation in those knowledge jobs. Will this be a painful revolution as the Industrial one? Where will we spend our time and energy on next?

> Where will we be in the future?

A German named Karl wrote a book with an answer to that question 150 years ago.

https://en.wikipedia.org/wiki/Das_Kapital

I shudder to think the direction this is all leading to.

The reason I'm a pessimist is that I mostly see society preserving the status quo. I don't see AI democratizing things and freeing us, because we have fetishized the concepts of work and profit that we can't imagine a society that functions properly without demanding those two things be put above all else.

I don't envision a dystopia or a utopia, I envision a future where AI disenfranchises people who should be taken care of, bolsters profits of the already powerful, and replaces the most fulfilling human pursuits without actually saving people from unfulfilling toil, mostly because society will bend backwards to try to preserve the status quo.

True art will be rare and treasured, few artists will be rockstars. Till next fad we can't even see coming again mixes it all up.

But yeah these transitions do make tons of people miserable, losing jobs. Also overall middle class is disappearing, but thats the trend for quite some time. But it always paid off to anticipate what next generation of showels will be.

They don't build 'em like they used to, but those of us in the rank and file are getting better off. We can't all be "I think I will just buy Hawaii" rich like Zuckerberg, but quality of life is, overall, improving.

I’ve decided the best thing I can do is vote with my wallet (i.e. as a consumer) and vote with my time (i.e. as a producer).

Consume commodities that are regenerative and products that are durable. Whenever possible, buy local to keep cash flows (wealth transfer) local.

Invest your time in contributing to products that encourage other people to do the same.

Yvon Chouniard said (I’m completely paraphrasing) something like “in the 21st century, the most powerful vehicle for transforming the world is the American corporation. Whether or not it should be, it is. If you think the world should be different, step 1 is starting a business”.

As a producer (engineer, for me), the best thing I can do is build tools to help people wean themselves off of Capital.

Buying (food, especially) locally means supporting a robust food chain that’s less dependent on refrigeration, global shipping, and a geographic arbitrage on agricultural labor. It’s better for the environment and connects you to your community and sense of place and seasons.

It means I can know what went into the food, in terms of soil, labor, and pesticides — something that I have no knowledge of at the grocery store.

A pound of apples, meat, or whatever from 1-2 counties away is more financially expensive, but if you can afford it there’s many non-financial benefits.

I’d rather purchase from a CSA than a national grocer, because that money mostly goes back into my local community.

You’re right the money doesn’t disappear when you spend it outside your locality, but it comes back in ways that are increasingly exploitative and/or depressing (chain stores offering no meaningful careers or opportunities for ownership, rent-seeking financialization, warehouses that bring noise pollution and packaging waste, etc.).

Again, if you can afford it, IDK why you wouldn’t prefer to spend the marginal dollar locally (i.e. private / family-owned businesses in a few counties’ radius).

It’s consumer choice and comparative advantage that actually lead people like me to shop locally...

> we keep destroying what allows humans to seek meaning.

> Our governments and society have made it clear, if you don't produce value, you don't deserve dignity.

Those points are really interesting to me, I think you're right and I've never looked at it that way before.

It's forcing the rich to give us UBI that is the problem.

It would be bare minimum to survive, if that. We'd need a complete restructuring of society to even approach dignity via UBI.

Nobody stops you from paying $1000 for a shirt made by artisans right now. Do you want to?

> Our governments and society have made it clear, if you don't produce value, you don't deserve dignity.

It's not somebody else who decided that. It's you.

If you don't do then your point is invalid. /s. Look, you dont have to pay $1000 for a shirt made by an artisan to get a quality shirt. It can be a bit more expensive, but much cheaper than that. And this can be true while being fair/reasonable to the artisan and also to the person acquiring it.

> It's not somebody else who decided that. It's you.

And what about my decisions? Or your decisions. Or anyone who is reading this's decisions? Surely the person you're replying to isn't some dictator who is deciding everything. It's not just the parent comment then. It's not just "It's you"

> If you don't do then your point is invalid. /s.

> Look, you dont have to pay $1000 for a shirt made by an artisan to get a quality shirt. It can be a bit more expensive, but much cheaper than that. And this can be true while being fair/reasonable to the artisan and also to the person acquiring it.

In the ideal case, this would be true: Following classical economics, there would be some predictable demand left as you go up the price curve, assuming genuine quality & labour.

However, this doesn't seem to be the case. The existence of low-cost mass-produced options leads to the satiation of demand above that spot of the curve. There still exists demand, but it's weirdly lower than predicted. People can shop down the price curve, but relatively fewer would do the opposite.

Aside from that I do think you make a good point and I'll need to think about it.

That was five years ago, with a recent re-sole this past summer.

This isn't an exaggeration by the way, but the cobbler who did the resole thanked me for bringing him the job, as it was a genuinely enjoyable experience for him. I assume most of his work is repairing suitcases going by the other clientele in the shop.

But sure, exaggerate that to be an artisan you need to be selling 1000 dollar shirts, rather than doing a 100 dollar service that doubles or triples the life of a decently made item.

> It's not somebody else who decided that. It's you.

No. The Republican National Convention cheered the idea of letting the poor and sick just die off in the streets. They cheered. Ron Paul asked "what are we supposed to do? Let our sick and poor just die cold in the streets?" and they cheered.

Jobs are sacred in the U.S. Job creators must be worshipped. Hard-working Americans are the lifeblood of yadda yadda. As soon as you don't have a job: fuck you, scum, you deserve to die in the streets. You are no longer of use to the wealthy, so you do not even have the dignity to sleep on benches or under bridges: they add spikes to any area you might find any comfort. Your children cannot receive an education. You get to disappear from view into some secluded slum until you die of the cold. It's not GP that decided that. It's tens (hundreds?) of millions of Americans who will cheer on your death if you lose your job. Most of which, of course, are a couple paychecks or a major illness away from being homeless themselves. Do not act like that's not a real thing.

He is one of a very small group of people (going back to 1989) to get a perfect raw score at the IoI, the olympiad for competitive programming.

https://stats.ioinformatics.org/people/2686

Glad to see that he's putting his (unbelievable) talents to use. To give you a sense, at the event where I met him, he solved 6 problems equivalent to Leetcode medium-to-hard problems in under 15 minutes (total), including reading the problems, implementing input parsing, debugging, and submitting the solutions.

If anyone is practicing for their B1 Dutch exam, feel free to use this link to get the practice paper.

https://usacognition--serve-s3-files.modal.run/attachments/4...

Even so, I realize the demos are still broad in scope and the results are incredible. Imagine seeing this even 2 years ago. It would seem like magic; you wouldn't be able to believe it. Today, this was inevitable and entirely believable. There will be even better versions of this soon.

In my mind, I've concluded that I have less than 3 years to find an off ramp.

I'm looking for an offramp as well. I truly love software so it's a hard reality to contend with at times. Regardless, I'm not a software engineer at my core. I'm a problem solver, and I love creating things. This is transferable. I'm not sure where to yet, but I'm certainly capable of making a move. I hope I have more than 3 years, though.

As someone else asked, I'm curious where you're headed or thinking of heading so far.

I think the immediate future is bright. Some will try to cram this technology into enterprise. It will do well. Those jobs will die. Others will leverage it alongside the more creative engineering tasks—they will thrive.

Eventually, what we call software will change from what it is today into something much more accessible to these types of tools. The plateau we've landed on is just a compromise between the technology, economies, and culture of its time.

As this type of tech pervades our everyday lives, much of the widespread need for specialized software will evaporate. The remaining work will be in the corners or the edges of what's possible—highly technical or vertically integrated work (particularly, hardware-integrated stuff), as well as platform engineering to sustain the new paradigm.

This change is roughly analogous to the shift from punch cards to compilers. It's just a more "natural language" way to do programming. A lot of the drudgery associated with coding will go away and competent programmers will shift to higher level design work.

Even in a future where AI is better than human software engineers at every single programming task (which I don't believe will be the case any time soon), there is still comparative advantage. AI will not have the _capacity_ to do every single programming task and there's going to still be lots of work for people to do.

Now everyone seems to be a "doomer" like me. I was too stupid to see that people (especially online) were lashing out at me as a coping mechanism. The worst part is that even though I had my "we're fucked" realization was earlier than many others, I didn't really do anything productive, I didn't make the right moves to pivot, I had no plan. Now I have a rough ~3-year off ramp plan, just like everyone else.

I suspect that there are many, many recent grads in my position, who also thought they were taking crazy pills when everyone around them dismissed their concerns

They don't know how the industry works or what it needs, few of them them know how the technology works or what it's realistic near-term prospects are let alone the long-term ones, and none of them have lived long enough to understand the pace of a technology moving from discovery through to maturity and commercialization.

If you look to your experienced seniors instead of your peers, you'll discover two things:

1. They don't respect most contemporary juniors because of the industry boom and are happy to see most of you scare yourselves off

2. They're narrowing in on a familiar, respectful appreciation of recent content generators and their potential for assisting some basic use cases (highly valuable technology!) and are increasingly of the opinion that both the hype and fear of Q1/2023 was largely market manipulation by VC's trying to squeeze another boom out as money was tightening

But something changed in the past 4 months or so. I used to see the two things you described, but now I see seniors repeating the same thing juniors (whom they ridiculed) used to say. Look at this very thread, for example. People are starting to realize that even if these AI startups are making ridiculous promises, the tech might be way more than just hype/smoke and mirrors.

I've read quite a few comments in HN, and right here in this thread, that say something along the lines of "I'll be fine... It's the juniors I'm worried about".

"Surely if we generate code, it means we can replace developers right?"

Well no, it's not nearly enough. Good code generation with a UI is already a thing since the 90s, it's replacing developers as much as Frontpage and WordPress do.

The average developer writes 100 lines per day anyways so the speed of writing those lines doesn't matter.

Software is an interpersonal job at core and the most precise definition of a business.

The day software engineers are replaced by a machine, there's no job for the CEO itself honestly and that's also why this job is well paid. I'm sure it will be all over LinkedIn though, despite this paradox.

Those new tools will be there in the background but cannot replace a software engineer.

I have a multi-part plan. Immediately, i'm working to get closer to the business. To be closer to the position of defining requirements, not implementing requirements.

Secondary, i'm experimenting with ideas I hope can become a business.

as a final fallback, I have a hobby woodshop in my basement, and I love making furniture.

Programmers will be fine. The AI plagiarism engines are severely limited, and will be for the forseeable future. Maybe someday i'll be equivalent to the ren-faire blacksmith, but I'm gonna do this until I die.

- deobfuscate complex requirements into well divided chunks

- find gaps or holes in requirements so that I have to write the minimal amount of code

- understand codebases so that the implementation fits nicely

I don't need an "AI software engineer", I need an "AI people person who gives me well defined tasks". Now sure, if you combine those two kinds of AIs I could perhaps become irrelevant.

It's funny how a year ago I was really excited for how AI can help my everyday coding while fearful that it would replace me. Now I'm not really sure either will happen in the short term.

Somewhat related: is anyone else not really impressed by Devin fixing the errors that are very preventable with a stricter language like Rust? The demo shows Devin coding in both Python in Rust, but I consider the latter being way less energy intensive in terms of maintenance. Then again, exhaustive pattern matching and strict typing won't get you lots of VC dollars these days.

Sure, one day we'll have AGI, and one day AGI will replace many jobs that can be done in front of a computer.

In the meantime, SOTA AI appears to be an airline chatbot that gets the company sued for lying to the customer. This is just basic question answering, and it can't even get that right. Would you trust it to write the autopilot code to fly the airplane? Maybe to write a tiny bit of it - just code up one function, perhaps?

I sure as hell wouldn't, and when it can be trusted to write one function that meets requirements and has no bugs, it's still going to be a LONG way before it can replace the job of the developers who were given a task of "write us an autopilot".

Most students who started jobs in software aren't writing software for aeroplanes. They're writing crud apps, doing third-party API integrations or doing basic debugging.

I worry that although not some specialized software devs but a lot will still have problems due to stuff like this.

I'm not talking today, but say 2-3 years down the line, who will have an intern when you can get an AI intern that can perform as the top percentile and comes at $20 per month subscription?

Currently, mainstream AI usage in coding is at the level of assistants and glorified autocomplete. Which is great (I use GitHub Copilot daily), but for us working in the space it's obvious that the impact will be much larger. Besides us (Pythagora), there's also Sweep (mentioned by others in the comments) and GPT Engineer who are tackling the same problem, I believe each with a slightly different angle.

Our thesis is that human in the loop is key. In coding, you can think of LLMs as a very eager junior developer who can easily read StackOverflow but doesn't really think twice before jumping to implementation. With guidance (a LOT in terms of internal prompts, and some by human) it can achieve spectacular results.

Why would would Agents need multiple turns and a framework like Pythagora in a (near-future) world with GPT4-level of capability and 10M+ token context?

Imagine GPT4 (or any other LLM) as a very eager but not very bright junior developer that just started to work with you. It's good, but it'll need a lot of situational management for it to not go wildly off the track.

What improved models will bring us in the future is making Pythagora and other such tools work better for large and more complex projects. The tipping point will come when for example you'll be able to load Pythagora in Pythagora and continue development. While we do build some auxiliary/external tools with Pythagora, the core is still handcrafted mostly by a human, and I'm pretty sure that's the case with other code gen tools as well.

Because it's better. This was the insight behind AlphaGo, AlphaCode and AlphaGeometry. In Go, trying to get the entire solution in one forward pass required a 1000x larger neural net to achieve the same performance according to Noam Brown.

Intelligence requires iteration, introspection, planning, sub-task breakdown, and trial and error with feedback from reality (error messages, print statements). These meta reasoning structures (whether hard-coded or, ideally, learned) will move LLMs closer to how humans reason.

I see LLMs failing at coding daily (one of the "perks" of working in the space), and I'm incredibly bullish on this approach.

And I don't think it'll replace humans or junior engineers. As programmers, we've been "replacing" ourselves since the days of assembler that replaced direct machine coding. This is just another iteration of it.

(if you do want a sales pitch, here's one: https://twitter.com/senkorasic/status/1765769482985722267 )

They'd better have really advanced reasoning and planning capabilities way beyond everything that anyone else knows how to do with LLMs. There's a growing body of literature that leaves no doubt that LLMs can't reason and can't plan.

For a quick summary of some such results see:

https://arxiv.org/pdf/2403.04121.pdf

Instead it will mean that bosses can fire 75-90% of the (very expensive) engineers, with the ones who remain left to prompt the AI and clean up any mistakes/misunderstandings.

I guess this is the future. We've coded ourselves out of a job. People are smiling and celebrating this all - personally I find it kinda sad that we've basically put an end to software engineering as a career and put loads of people out of work. it is not just SWEs - it is impacting a lot of careers... I hope these researchers can sleep well at night because they're dooming huge swathes of people to unemployment.

Are we about to enter a software engineering winter? People will find new careers, no kids will learn to code since AI can do it all. We'll end up with a load of AI researchers being "the new SWEs", but relying on AI to implement everything? Maybe that will work and we'll have a virtuous circle of AIs making AI improvements and we'll never need engineers again? Or maybe we'll hit a wall and progress in comp sci will essentially stop?

Secondly, bosses try to fire the most expensive programmers ever since expensive programmers became a thing. All the outsourcing to wherever the salaries are much smaller is being attempted even as I type this now, by probably no less than 1000 companies, in this very second. Why hasn't the area in general still gotten rid of their expensive programmers?

Easy -- the outsourced "talent" produces crap that then costs more to fix and repair than if you hired proper programmers in the first place. Of course that requires some forward thinking that's not limited only by the short-sighted "we are about to save money muahahaha" mindset and as we know many businessmen are incapable of looking forward -- hence this mistake is being done 24/7.

It's 99% the same with these AI coding bots.

Wishing cost savings into existence so far hasn't worked.

Will some interns and smart people who coast on much-higher-than-local wages because they wrote some Python to import their boss' Excel spreadsheets and automatically generate other stuff, lose their jobs? Yes, that's very likely.

Will the senior programmer be replaced? It's not a zero chance, surely, and we already saw significant layoffs in most big US companies but (1) most of the world didn't over-hire during COVID and (2) it's unclear if senior devs were fired, or actual redundancies, and (3) well, most of the world isn't the USA.

So again, you are vastly overselling the current generation AI.

Whether or not the tech can actually live up to the hype doesn't mean execs/VCs/etc won't try to get there anyway. That will, at best, result in a ton of volatility, as those trying to utilize AI figure out that it actually can't do what they want, and then have to hire engineers again, etc...

I don't know. Who knows where this is heading.

We continually hear anxiety about technology leading to mass unemployment and it keeps not happening. Instead, workers tend to have higher productivity, which drives higher wages.

Technological advancements transformed agriculture from being 75% of the workforce to being less than 5% of the workforce over the last 200 years and instead of mass unemployment, everyone found other ways to add value to society and it has been an absolute win, with higher standards of living.

(https://www.researchgate.net/figure/Percent-of-the-Labor-For...)