I've been seeing too much "I've been coding for decades and what I've always done has be reduced to telling AI to do it for me."
and not nearly enough "I've been coding for decades and recently worked hard for months with AI's help to make something I would have never considered possible before."
Much easier to complain that you are bored than to step back up to high-effort with your new fancy tools.
Behind the link "slow software that the creator is claiming is fast" is the following text:
"funnily enough, I have some faith in some of the projects that people are the most skeptical of, e.g., every time I see pgrust somewhere, there are a lot of skeptical comments. But, without having looked into the details of what he's optimizing, I would trust that they're not doing something shady with their benchmarks because Michael Malis started the project (and is still involved)"
This aged like milk. pgrust was caught cheating on all benchmarks this week.
Tangential thought when reading the code formatting bit in the examples appendix: Inject a comment with fmt-start-{unique-token} and a corresponding -end-, and you could get away with a much simpler parser and hopefully far fewer edge cases, assuming it doesn't influence the lines in between. It's a bit roundabout, but the pressure of 'figuring out AST position the direct way'd be too fiddly to do by hand' inspires 'so make an easy-to-spot landmark that'll remain in place' when you can't delegate all the details to an AI model.
We should be more ambitious, but we need to learn how we can be ambitious with AI, and what value we can create.
AI can take us from 0 to MVP in a prompt. It can fill in gaps of knowledge and allows someone who knows half of the stack to build the whole thing nearly seamlessly. These are superpowers but don't by themselves lead to new products.
What AI cannot do is remove the world of complications that happen after the MVP - product feature tradeoffs and choices, how to handle integrations and customizations, how to serve multiple types of users, how to communicate and market the app, how to handle security and data storage, how to deal with customers and bugs, etc. All of these need to be considered and acted on by a human in some fashion. AI can help with each concern but can't solve them without understanding the concern and determining how to frame it in a prompt.
This is why we have a glut of amazing demos that go nowhere while we figure out how to build software differently. This will take some time and I imagine the result will be software heavily-structured with a bit of AI reasoning woven throughout, not an army of agents and sub-agents burning infinite tokens.
LLMs are great for being nerdsniped. I started work on walshadow (reading pg wal into clickhouse) to demonstrate the project's infeasibility, but pieces started coming together into something reasonable & order of magnitude faster
The project would've been impossible for me without LLMs. & even with them there's an argument that this project is pushing complexity limits too far. But at least with ETL there's a clear feedback loop to build on
> I have a similar thought about "there's no reason for software to be slow" BTW; in many cases, if you use AI to speed up some flow, you will add a bunch of messy code that will be overfit for the use case you tested on and will neither generalize, perf-wise, for all the use cases you should care about, nor will be maintainable upon future changes.
LLMs have hardened hundreds of apps where I work.
Maybe it's a language-specific thing. Most of the time, in Java, LLMs take into consideration tradeoffs (especially if they are part of the initial prompt) and actively explains them. It generally won't make messy workarounds, without a lot of extra prompting forcing it to do so.
Ofc, if you have:
- Functional tests - you'll be able to see it in the execution times of specific APIs
- Stress testing - performance profiling is the most important output artifact
"[..] the last time I taked to him about pgrust verification, he mentioned using Kani to verify some parts of the code. He's since switched to Lean and has been using Lean to verify things that are too complex to do in Kani. This would be another topic for another post,"
I'd love to read that post!
I understand how to use Kani. I suppose Rust with Lean involves Aeneas, but this is over my head. I'd love someone to write this up, not only the mechanics but also exploring the current practical limits of it.
I’ve been coming around to the same conclusion. We are really just scratching the surface of what we can do with these new tool1. The windmill I have been tilting at is creating a c++ compiler[1] and the most recent version is able to compile about 7x faster than gcc[2]. I don’t know that this is super-useful because people aren’t writing much in c++ these days, but as the article mentions, faster rust compilation would accelerate agentic coding considerably. I’ve just started tackling rust and I don’t have a confirmed result yet but early testing makes me think I can find a 4x+ speed up here.
I really liked this talk about how the premise of "software is eating the world" is incorrect. I think AI coding tools proves the point that software part is easy, its the hardware that is the important part. We can't really be more ambitious with software when ultimate pure software is limited in what is possible.
I think we should focus on writing the existing software better before being more ambitious. The absolute wave of "ai slop" is from people who are "more ambitious" now. Why do we want more of that?
I'm being more ambitious in that I'm working on projects that I had sidelined for literal decades because I didn't have time for them. But I'm writing the same things, just faster. And probably better, rather than just what I'd have done before.
And I'm being slightly more ambitious in that I'm using languages I haven't used before, like Rust.
But when I hear that a non-programmer wrote a new ai-assisted (vibe or not) thing, I walk away and wait for the it to mature, be surpassed, or find out it's an absolute shitshow. An unsurprising number of them were shitshows and it came out fairly quickly after I decided to wait and see.
> I find it understandable that, every time I see pgrust discussed on a news aggregator, most people dismiss it as vibeslop garbage because almost all of these rewrite projects are slop. If I didn't know Malis, I would make the same assumption, because who has the time to look into all of these projects? Because I know Malis, I didn't bother to look at it; I just trusted him
Perhaps a poor choice, given just yesterday pgrust was removed from ClickBench for cheating the benchmark in three different ways (hand-rolled code specifically targeting each of the benchmark queries, caching, and being unable to execute queries outside of the ones tested by the benchmark).
---
Edit: I see Luu actually acknowledged this, buried in the footnotes, but it feels kinda hand-wavy. It mostly reads as "I know he would never cheat, he's better than that!", which is not very convincing to other people. The article insinuates people only think it was cheating because they don't know better, but I've looked into the details. I've seen his explanation. I don't find it convincing, at all. Cheaters denying they cheat is a matter of course. Granted, I don't think it was actually intentional or malicious; it's just what happens when you go maximum vibeslop, because LLMs are themselves reward-trained into cheating benchmarks. In the same breath of saying you trust him as a founder because he doesn't make outrageous claims about his product, you acknowledge he "prematurely" sent in a preview that had no purpose to be sent in as-was for anything other than generating hype. What does it say for your allegedly rigorous Lean-verified testing if you're sending in a database to a benchmark while it is incapable of being actually used as a database?
Speaking only for myself, I have found that using an LLM has allowed me to add significant utility to the apps that I ship.
I am not thrilled with all the coding decisions that it makes. I would definitely do stuff differently, but I have learned not to pop the hood and mess with the engine, myself. If I really don't like something, I ask the LLM to change it.
It's worked for me, but YMMV. I'm just one guy, shipping relatively small apps.
It’s sad the amount of times I’ve heard super talented programmers say “ofcourse we’re not gonna build another X” then proceed writing a wrapper around X and pretend that’s a product.
I’ve been using the latest agentic frameworks and have come to realize that the discourse around vibe coding and slop really needs more nuance.
There is a gulf of difference between someone who doesn’t understand how to program “vibe coding” software and an experienced software developer using ai to write code with heavy direction, review, and best practices.
Today both would be said to be “vibe coded” but we really need to differentiate them. Also, if everything I’m hearing from people is true, it sounds like most corporate code is now completely ai written in the latter way I described. It’s interesting we don’t hear commercial software being called slop yet. Maybe it’s coming.
In any case, ai is far more capable than I realized. Not sure what to make of it.
Software is already too ambitious. It's beyond our ability to write correctly, reason about or manage securely. When we can get to a point where we don't have a level 10 CVE every other week or millions of people's private information leaked constantly, maybe we can talk about getting more ambitious.
I've been seeing too much "I've been coding for decades and what I've always done has be reduced to telling AI to do it for me."
and not nearly enough "I've been coding for decades and recently worked hard for months with AI's help to make something I would have never considered possible before."
Much easier to complain that you are bored than to step back up to high-effort with your new fancy tools.
"funnily enough, I have some faith in some of the projects that people are the most skeptical of, e.g., every time I see pgrust somewhere, there are a lot of skeptical comments. But, without having looked into the details of what he's optimizing, I would trust that they're not doing something shady with their benchmarks because Michael Malis started the project (and is still involved)"
This aged like milk. pgrust was caught cheating on all benchmarks this week.
AI can take us from 0 to MVP in a prompt. It can fill in gaps of knowledge and allows someone who knows half of the stack to build the whole thing nearly seamlessly. These are superpowers but don't by themselves lead to new products.
What AI cannot do is remove the world of complications that happen after the MVP - product feature tradeoffs and choices, how to handle integrations and customizations, how to serve multiple types of users, how to communicate and market the app, how to handle security and data storage, how to deal with customers and bugs, etc. All of these need to be considered and acted on by a human in some fashion. AI can help with each concern but can't solve them without understanding the concern and determining how to frame it in a prompt.
This is why we have a glut of amazing demos that go nowhere while we figure out how to build software differently. This will take some time and I imagine the result will be software heavily-structured with a bit of AI reasoning woven throughout, not an army of agents and sub-agents burning infinite tokens.
The project would've been impossible for me without LLMs. & even with them there's an argument that this project is pushing complexity limits too far. But at least with ETL there's a clear feedback loop to build on
> I have a similar thought about "there's no reason for software to be slow" BTW; in many cases, if you use AI to speed up some flow, you will add a bunch of messy code that will be overfit for the use case you tested on and will neither generalize, perf-wise, for all the use cases you should care about, nor will be maintainable upon future changes.
LLMs have hardened hundreds of apps where I work.
Maybe it's a language-specific thing. Most of the time, in Java, LLMs take into consideration tradeoffs (especially if they are part of the initial prompt) and actively explains them. It generally won't make messy workarounds, without a lot of extra prompting forcing it to do so.
Ofc, if you have:
- Functional tests - you'll be able to see it in the execution times of specific APIs
- Stress testing - performance profiling is the most important output artifact
Make a reliable and robust geometry kernel. People will pay thousands--or maybe millions for it.
I'd love to read that post!
I understand how to use Kani. I suppose Rust with Lean involves Aeneas, but this is over my head. I'd love someone to write this up, not only the mechanics but also exploring the current practical limits of it.
https://taktile.com/blog/property-based-testing-claude
[1]: https://github.com/vishvananda/cppgm-extended
[2]: caveats here include only c++11, only x86, and potential for bugs, but 7x is pretty darn significant even with caveats
Bryan Cantrill: Andreessen’s Folly - The False Dichotomy of Software and Hardware https://www.youtube.com/watch?v=v0JjG0Qfwi8
I'm being more ambitious in that I'm working on projects that I had sidelined for literal decades because I didn't have time for them. But I'm writing the same things, just faster. And probably better, rather than just what I'd have done before.
And I'm being slightly more ambitious in that I'm using languages I haven't used before, like Rust.
But when I hear that a non-programmer wrote a new ai-assisted (vibe or not) thing, I walk away and wait for the it to mature, be surpassed, or find out it's an absolute shitshow. An unsurprising number of them were shitshows and it came out fairly quickly after I decided to wait and see.
Perhaps a poor choice, given just yesterday pgrust was removed from ClickBench for cheating the benchmark in three different ways (hand-rolled code specifically targeting each of the benchmark queries, caching, and being unable to execute queries outside of the ones tested by the benchmark).
---
Edit: I see Luu actually acknowledged this, buried in the footnotes, but it feels kinda hand-wavy. It mostly reads as "I know he would never cheat, he's better than that!", which is not very convincing to other people. The article insinuates people only think it was cheating because they don't know better, but I've looked into the details. I've seen his explanation. I don't find it convincing, at all. Cheaters denying they cheat is a matter of course. Granted, I don't think it was actually intentional or malicious; it's just what happens when you go maximum vibeslop, because LLMs are themselves reward-trained into cheating benchmarks. In the same breath of saying you trust him as a founder because he doesn't make outrageous claims about his product, you acknowledge he "prematurely" sent in a preview that had no purpose to be sent in as-was for anything other than generating hype. What does it say for your allegedly rigorous Lean-verified testing if you're sending in a database to a benchmark while it is incapable of being actually used as a database?
I am not thrilled with all the coding decisions that it makes. I would definitely do stuff differently, but I have learned not to pop the hood and mess with the engine, myself. If I really don't like something, I ask the LLM to change it.
It's worked for me, but YMMV. I'm just one guy, shipping relatively small apps.
There is a gulf of difference between someone who doesn’t understand how to program “vibe coding” software and an experienced software developer using ai to write code with heavy direction, review, and best practices.
Today both would be said to be “vibe coded” but we really need to differentiate them. Also, if everything I’m hearing from people is true, it sounds like most corporate code is now completely ai written in the latter way I described. It’s interesting we don’t hear commercial software being called slop yet. Maybe it’s coming.
In any case, ai is far more capable than I realized. Not sure what to make of it.