nevertoolate

pgrust is ambitious but shall every developer build their postgres to prove they are good enough developers? The project claims to "show what Postgres would look like if it was built in 2026". It shows the traits yes: a single 3M line commit; no architecture discussion; no lessons learnt; claude it he top contributor; people writing emotional articles about why such projects are good.

It makes it really hard to connect to any of this. I loved danluu articles before, but somehow they've started to drift into this hard to grasp, hard to argue against shallow and empty world I try to get away from.

Interesting that you somehow start to talk about religion. I don’t think having a hypothesis about an eternal soul makes someone religious. Believing in “technology” or “technological advancement” seems to be a similar hypothesis as we clearly don’t see the future. The thing is that memories, input peripherals, the whole body is usually out of our conscious understanding so it seems to be a pretty farfetched idea that we operate in world where we don’t need pretty wild hypothesis about kinda everything without almost zero facts to depend on. Blue pill / red pill

It looks like a next token predictor, walks like a next token…

You get my point. It definitely doesn’t look like my elderly neighbour, nor like my daughter, etc. It is confusing but very simple at the same time.

> Approaching it from first principles means starting with what we’re trying to do and asking how AI could help

I’m not sure why would you want to think about how an AI would help. If you understand what you need to do and you understand the capabilities of the tools available you run the coding / learning loop. If you think you should use AI then you are solving the how to use AI problem, which just adds to the complexity. Orthogonal to first principle thinking, just a workflow issue.

Is this a loaded question? People are getting hooked by high(ish) quality shows for relatively low prices. Once hooked you need to make them binge content (algorithm). Once hooked you can sell the ** for higher price.

You know what to do. Just don’t watch shit :)

It is called micromanagement when the boss tells you _how_ you should think. My approach is when leading a tech team to find together the things which should be done and find a method together to keep us on the path. Facilitation. Forcing decisions to be made. Keep quality high enough. Let people find what they enjoy doing and is useful. This is really basic and well known set of tools.

Quick feedback: this framework feels like a not so well thought out extension of polya’s how to solve it to “do things fast” in a credible looking way.

Harder feedback: the writeup feels like a quickly hacked together memo, which claims it is “self healingly” good. Because it has this throwaway quality I’m not sure what is the goal? Shall I engage with it seriously, when the lead dev didn’t seem to put in the effort? But if I don’t engage with it seriously it means that I have to follow this patchy fremwork?

The author is trying to codify, and partially automate, a pretty common pattern to create crud style apps. Or what do you think they propose?

So you want to reduce entropy by “fixing” semantic mistakes and hardcoding some design decisions “by the book” and end up with what exactly? The exact same poster everywhere?

I know this is an internet forum, but do you have any experience with posters? There are many great tools to create posters from templates. You can make something in canva for example in minutes from free templates. You upload a photo, play around with it. You can also search the old internet for some tutorials. You know, like a human would do.

Prompt1: generate random poster

Polish: make it nicer

I have aphantasia, figured out after a vivid dream which I’ve told my wife about so happily. She was not really amused. She sees these things ALL THE TIME. Like whenever she wants, or from time to time even when she doesn’t want to. She saw his father on the street one day. Two months after he had died.

We have a hard time communicating about things. She sometimes can’t express herself at all, she sees things she thinks I should be able to see. I have very good ability to remember schemata, maps. I move really quickly in pages and pages of text. I can summarize the pages and pages of llm output in less time than it is generated. She can easily put 8 grammatical mistakes in a 5 word sentence. We can communicate easily when i write the code quickly and she can checks the visual output in the browser, this is really fluid.

Want more examples ? :)

So now you have two vibe coded applications you don’t understand. I’m not an advocate of making “job security” decisions but this definitely goes to red flag territory. Was it your job to maintain the RN codebase or do you have other functions there as well? I’m not sure I would keep a native noob on a vibed native codebase. What is your take?

I’m not sure who will be solving these incidents if 99% of incidents will be “solved” by LLMs. If I’m called once a year my daily rate will be my yearly rate?

> For example, a worker taking on a new position now has access to a tutor and an analyst and a business advisor all wrapped up in one

These llm agents don’t have “agency”. No intrinsic goal other than respond. A tutor has judgment and agency. You will get a really good pattern match on the input text you give but you won’t get tutoring.

> Let me get this straight. You think that your client, one of the wealthiest, most powerful men in the world, is secretly a vigilante who spends his nights beating criminals to a pulp with his bare hands...and your plan is to blackmail this person? [pauses, smiles] Good luck.

Are you an AI?

I like this story. How did you verify the output? How big is the codebase? Why it took 18 hours? Could you implement it with a small local agent and breaking the task down yourself in two days (i know it sounds like a loaded question, it is not).

I think the rewrites are the main story for llms in code (hot take). Writing greenfield code at the seams also something which might work well.

They didn’t really sell this PR well:

> Workflows like jira_close.yml use deprecated atlassian JIRA actions and have a dependency on the gh-actions repo. This is not ideal and unecessarily complex.

And then goes on:

> PR updates jira_close workflow to use direct API calls via curl.

Duplicating the logic into OUR codebase via a hand rolled curl, so we can get rid of “needless abstractions”. Auch. And of course the whole thing embedded into a yaml file.

This code is the typical kaleidoscope sometimes written by junior devs (and LLMs). On review you just kindly ask to be rewritten into a simple program or just close it as the effort doesn’t worth it.

> [sota llms] still being not smart enough to refactor large codebases as well as me, so it’s hard to imagine a world where I don’t just want to use the smartest model available

I think this is the core point of the article. I really hope that we will stop prompting llms to refactor codebases and start using deterministic llm supported toolchains for such tasks. I think you either blindly trust an llm or you don’t. And no post-hoc review will be enough to contain llms, we will super quickly create “dark factory” without the benefits.

Have you realized you are attacking a straw-man? Nobody said all “hand written” code is good. I _think_ the implicit argument is: “all vibed code lacks cohesion”.

Wait you only wanna make llm write the boring parts but after it has written it you retype. Without concrete example this sounds bizarre.

There is a glaring fallacy in your “AI will change everything as it is super intelligent” hypothesis. If it is so great thinker which can do everything why not just solve this social impact thingie? Or maybe it is not so capable?

Css classes are not big issue with css nesting and if component boundaries are settled and the design is well defined before implementation. Tailwind is a shorthand notation which is harder to parse than a structured css file (at least I strongly believe this we can agree on) and thus has only editing speed gains which is largely not an issue if prior work is done on exploration. Another way to frame it: tailwind is a good prototype syntax but is not the best syntax to for production code (human parsability). I’d skip it altogether to simplify the flow.

Right so it is easier to write code quickly, but not to maintain a system. Tailwind is the right tool if you want to generate a prototype with LLM agents. It can be quickly deleted or can be used as visual prototype to create reusable components. And once the component is built for long term maintenance as priority, the collective knowledge has been built up so naming becomes simpler, important and elucidating.

> We’re required to use an LLM for every task at work that touches code

Why not just write the code in the prompt so LLM can paste it.

If it is a game of survival yes, we are living on the edge. Is it.

If you live your life in stress it makes it strange. Breathe less, stress less.

> I had to scroll past so many cynical cheap shots

I haven't read cynical comments, just ones pointing out that the article is cynical itself.

> addresses the core point

No it doesn't address anything, it is fear mongering question.

> I have yet to hear a single compelling plan for how we will prevent bioweapon development or massive hacking campaigns

Me neither, I just see marketing campaigns trying to raise valuation of a pre-IPO company.

> you need to suggest an alternative plan that addresses these concerns

I suggest that Dario stops writing marketing letters and start organizing a mostly neutral expert organization to propose solutions.

Also notice that as EU citizen I don't trust a US pre-IPO company's CEO with conflict of interest to suggest solution on resolving global security matters. Especially since he admittedly has no control over how the technology of his own company is deployed in global conflicts[1]

[1] https://www.forbes.com/sites/antoniopequenoiv/2026/06/10/ant...