HNHacker News
TopNewBestAskShowJobs

rudedogg

2,349 karma · joined February 15, 2013

submissionscomments
rudedogg··on Switch to Claude without starting over
I've been thinking this too. I frequently do deep research on some systems programming technique, ask it to generate a .md for it, and then I use that in later sessions with Claude Code "look at the research I collected in {*-research}.md and help me explore ways to apply it to {thing}".

At the research step it frequently (always?) uses memory to direct/scope the research to what I typically work on, but I think that kind of pigeon holes the model and what it explores. And the memory doesn't quite capture all the areas I'm interested in, or want to directly apply the research to.

And regarding the crap in memories, I found the same. Mine at work mentioned I'm an expert at a business domain I have almost zero experience with.

I feel like the companies building this stuff accept a lot of "slop" in their approach, and just can't see past building things by slopping stuff into prompts. I wish they'd explore more rigid approaches. Yes, I understand "the bitter lesson" but it seems obvious to me some traditional approaches would yield better results for the foreseeable future. Less magic (which is just running things through the cheapest model they have and dumping it in every chat). It seems like poison.

Related: https://vercel.com/blog/agents-md-outperforms-skills-in-our-...

Also, agent skills are usually pure slop. If you look through https://skills.sh on a framework/topic you're knowledgeable in you'll be a bit disheartened. This stuff was pioneered by people who move fast, but I think it's now time to try and push for quality and care in the approach since these have gotten good enough to contribute to more than prototype work.

rudedogg··on Our Agreement with the Department of War
It's not much but I was planning to cancel my Anthropic subscription to try Codex over the weekend, but I'll skip that. I don't want to support a company with someone like this at the top. Massive donations to the administration, sneaky backdoor deals. No thanks, fuck you.
rudedogg··on Google workers seek 'red lines' on military A.I., echoing Anthropic
> And if China gets AI, they're more than likely to use it to further raise people out of poverty and automate away more menial jobs without making those displaced workers homeless.

Your comment is very optimistic. But the quoted part reminded me of something I heard (again) about China using slave labor in their lithium mines:

https://www.state.gov/forced-labor-in-chinas-xinjiang-region...

rudedogg··on Zig Libc
Thanks! I think I spent as much time writing the post as I did making the skill, so I’m happy someone got some value out of it.
rudedogg··on Zig Libc
A little bit! I wrote a long blog post about how I made it, I think the strategy of having an LLM look at individual std modules one by one make it actually pretty accurate. Not perfect, but better than I expected
rudedogg··on Zig Libc
I made a Zig agent skill yesterday if interested: https://github.com/rudedogg/zig-skills/

Claude getting the ArrayList API wrong every time was a major reason why

It’s AI generated but should help. I need to test and review it more (noticed it mentions async which isn’t in 0.15.x :| )

rudedogg··on Deep dive into Turso, the “SQLite rewrite in Rust”
From what I’ve read there’s a pretty sizable performance gap between SQLite and pglite (with SQLite being much faster).

I’m excited to see things improve though. Having a more traditional database, with more features and less historical weirdness on the client would be really cool.

Edit: https://pglite.dev/benchmarks actually not looking too bad.. I might have something new to try!

rudedogg··on AI Tribalism
This is kind of where I'm at.

I don't think everything is for certain though. I think it's 50/50 on whether Anthropic/whoever figures out how to turn them into more than a boilerplate generator.

The imprecision of LLMs is real, and a serious problem. And I think a lot of the engineering improvements (little s-curve gains or whatever) have caused more and more of these. Every step or improvement has some randomness/lossiness attached to it.

Context too small?:

- No worries, we'll compact (information loss)

- No problem, we'll fire off a bunch of agents each with their own little context window and small task to combat this. (You're trusting the coordinator to do this perfectly, and cutting the sub-agent off from the whole picture)

All of this is causing bugs/issues?:

- No worries, we'll have a review agent scan over the changes (They have the same issues though, not the full context, etc.)

Right now I think it's a fair opinion to say LLMs are poison and I don't want them to touch my codebase because they produce more output I can handle, and the mistakes they make are too subtle that I can't reliably catch them.

It's also fair to say that you don't care, and your work allows enough bugs/imprecision that you accept the risks. I do think there's a bit of an experience divide here, where people more experienced have been down the path of a codebase degrading until it's just too much to salvage – so I think that's part of why you see so much pushback. Others have worked in different environments, or projects of smaller scales where they haven't been bit by that before. But it's very easy to get to that place with SOTA LLMs today.

There's also the whole cost component to this. I think I disagree with the author about the value provided today. If costs were 5x what they are now, I think it would be a hard decision for me to decide if they are worth it. For prototypes, yes. But for serious work, where I need things to work right and be reasonably bug free, I don't know if the value works out.

I think everyone is right that we don't have the right architecture, and we're trying to fix layers of slop/imprecision by slapping on more layers of slop. Some of these issues/limitations seem fundamental and I don't know if little gains are going to change things much, but I'm really not sure and don't think I trust anyone working on the problem enough to tell me what the answer is. I guess we'll see in the next 6-12 months.

rudedogg··on Show HN: Agent Skills Leaderboard
My experience with them is limited, but I’m having issues with the LLMs ignoring the skills content. I guess it makes sense, it’s like any other piece of context.

But it’s put a damper in my dream of constraining them with well crafted skills, and producing high quality output.

rudedogg··on Apple testing new App Store design that blurs the line between ads and results
Yes, this is part of what is supposed to justify the premium prices, is that they can have a different business model.

But it seems Tim Cook can’t leave anything on the table. I’m really going to be irritated if we end up with a premium Siri. It’s going to undermine the privacy aspect, the hardware innovation, and everything else they have going for themselves despite missing the boat on AI

rudedogg··on The recurring dream of replacing developers
If these things can ever actually think and understand a codebase this mindset makes sense, but as of now it's a short-sighted way to work. The quality of the output is usually not great, and in some cases terrible. If you're just blindly accepting code with no review, eventually things are going to implode, and the AI is more limited than you are in understanding why. It's not going to save you in it's current form.
rudedogg··on The recurring dream of replacing developers
> increasing the level of abstraction developers can work at

Something is lost each step of the abstraction ladder we climb. And the latest rung uses natural language which introduces a lot of imprecision/slop, in a way that prior abstractions did not. And, this new technology providing the new abstraction is non-deterministic on top of that.

There's also the quality issue of the output you do get.

I don't think the analogy of the assembly -> C transition people like to use holds water – there are some similarities but LLMs have a lot of downsides.

rudedogg··on We put Claude Code in Rollercoaster Tycoon
LSP also kind of sucks. But the problem is all the big companies want big valuations, so they only chase generic solutions. That's why everything is a VS Code clone, etc..

https://paulgraham.com/ds.html

rudedogg··on The insecure evangelism of LLM maximalists
Hearing people on tech twitter say that LLMs always produce better code than they do by hand was pretty enlightening for me.

LLMs can produce better code for languages and domains I’m not proficient in, at a much faster rate, but damn it’s rare I look at LLM output and don’t spot something I’d do measurably better.

These things are average text generation machines. Yes you can improve the output quality by writing a good prompt that activates the right weights, getting you higher quality output. But if you’re seeing output that is consistently better than what you produce by hand, you’re probably just below average at programming. And yes, it matters sometimes. Look at the number of software bugs we’re all subjected to.

And let’s not forget that code is a liability. Utilizing code that was “cheap” to generate has a cost, which I’m sure will be the subject of much conversation in the near future.

rudedogg··on We can't have nice things because of AI scrapers
Also iCloud Private Relay.

CloudFlare is making it impossible to browse privately

rudedogg··on What happened to WebAssembly
> JIT compilation has the potential to exceed native compiled speeds

The battlecry of Java developers riding their tortoises.

Don’t we have decades of real-world experience showing native code almost always performs better?

For most things it doesn’t matter, but it always rubs me the wrong way when people mention this about JIT since it almost never works that way in the real world (you can look at web framework benchmarks as an easy example)

rudedogg··on Bankruptcies are exploding across the economy
I’ve been thinking about this a lot lately and agree.

Also, it’s amazing how inefficient medium+ businesses are. I think we should see small businesses thriving due to the cost/weight of the bureaucracy they inflict on themselves, but we don’t.

I think healthcare costs/requirements and unfair access to capital keep the inefficient machine chugging along, cutting off routes for smart people to start businesses, innovate, and improve our economy.

rudedogg··on Ensuring a National Policy Framework for Artificial Intelligence
I’m saying if the law was respected at all this technology would be banned. I don’t know that I prefer that outcome, but it is the truth.

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

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

rudedogg··on Ensuring a National Policy Framework for Artificial Intelligence
> Where did you get the idea that banning new technology that could eliminate jobs is even remotely an American value?

Copyright law is another counter-example to your argument. But somehow? that’s no longer a concern if you have enough money. I guess the trick is to steal from literally everyone so that no one entity can claim any measurable portion of the output as damages.

I’ve always thought Copyright should be way shorter than it is, but it’s suspect that we’re having a coming to Jesus moment about IP with all the AI grifting going on.

rudedogg··on Icons in Menus Everywhere – Send Help
Here's from 2020: https://web.archive.org/web/20201027235952/https://developer...

> Use text, not icons, for menu titles. Only menu bar extras use icons to represent menus. See Menu Bar Extras. It’s also not acceptable to use a mixture of text and icons in menu titles.

> Avoid using custom symbols in menus. People are familiar with the standard symbols. Using nonstandard symbols introduces visual clutter and could confuse the user.

The notable thing here is how recent of a shift this is, and how longstanding the prior rule was. Navigating internet archive is slow/tedious, but I think the rule/guideline was explicitly called out in the guidelines up until a year or two ago. So it was probably the guideline for ~20 years on macOS and has just now been changed.

rudedogg··on Has the cost of building software dropped 90%?
The author teaches AI workshops. Nothing wrong with that, but I think it should be disclosed here. A lot of money is riding on LLMs being financially successful which explains a lot of the hype.
rudedogg··on Why we built Lightpanda in Zig
Just Zed (if AI features are a requirement) as far as I know.

But to elaborate, they’ve found a niche simply by using Rust and rendering the GUI in a performant way on the GPU. I’m not saying performance is the only thing, but for a chunk of people it is something they care about.

rudedogg··on Why we built Lightpanda in Zig
The language tends to affect everything, but to give a quick Developer example there’s Zed. Developers use it because it’s fast. Same with Sublime Text.

Your criticism makes more sense with products targeting non-technical users though. But IMO tech choices have cascading effects. I won’t buy a vehicle if the infotainment software sucks, and that’s the 2nd largest purchase I’ll ever make.

rudedogg··on Cloudflare outage on December 5, 2025
I’m really sick of constantly seeing cloudflare, and their bullshit captchas. Please, look at how much grief they’re causing trying to be the gateway to the internet. Don’t give them this power
rudedogg··on Paul Hegarty's updated CS193p SwiftUI course released by Stanford
I'm thankful these are given away for free, but a little sad all the videos are only in 720p
rudedogg··on Pocketbase – open-source realtime back end in 1 file
> Could you elaborate? - thanks

The part that prompted my comment was clicking to edit a row on mobile in vertical orientation. Rather than fill the screen the edit shows up in a modal which doesn't really make sense, since you have so little horizontal space to work with. And the inputs are cut off slightly due to no padding.

And the whole admin UI overfills my (pro max) iPhone when held vertically, I think due to the navbar icons? Anyway, it'd be nice to not have to horizontally scroll to see those 40px or w/e everywhere. It makes it so you have to horizontally scroll to see or evaluate every page

One other little nice thing you could do if you wanted is prefill the username/password with the demo user.

---

Very cool project, and it's cool to see how Rust benefits the performance.

rudedogg··on Pocketbase – open-source realtime back end in 1 file
I appreciate their honesty. After a quick look I’d give another point to Pocketbase for it’s admin UI. The TrailBase one is pretty sloppy (on mobile at least), and looks like it’s using bootstrap.

Pocketbase has a sense of quality/care around it that seems missing.

rudedogg··on How I learned Vulkan and wrote a small game engine with it (2024)
RealityKit is pretty cool and the replacement it seems. Still Apple only though, and I find the feedback loop slow/frustrating due to Swift

I find SDL3 more fun and interesting, but it’s a ton of work to to get going.

rudedogg··on Gemini 3
Every recent release has bumped the pricing significantly. If I was building a product and my margins weren’t incredible I’d be concerned. The input price almost doubled with this one.
rudedogg··on Open-source Zig book
> Are you sure?

There are too many things off about the origin and author to not be suspicious of it. I’m not sure what the motivation was, but it seems likely. I do think they used the Zig source code heavily, and put together a pipeline of some sort feeding relevant context into the LLM, or maybe just codex or w/e instructed to read in the source.

It seems like it had to take quite a bit of effort to make, and is interesting on its own. And I would trust it more if I knew how it was made (LLMs or not).

As another suspicious data point see this issue by the author: https://github.com/microsoft/vscode/issues/272725

Edit: https://news.ycombinator.com/item?id=45952581 found some concrete issues

← PreviousPage 3 of 20Next →