HNHacker News
TopNewBestAskShowJobs

p1necone

10,610 karma · joined February 21, 2018

submissionscomments
p1necone··on Reducing undefined behavior in the C language
Hence the 'specific to C/C++' part of that sentence. 'Undefined behaviour' is a probably unavoidable part of programming language design. The way that C/C++ interpret/handle it is batshit insane.
p1necone··on Reducing undefined behavior in the C language
The concept of undefined behaviour specific to C/C++ has always seemed batshit insane to me, and I'm yet to read anything about it that has made it seem any less so.
p1necone··on Why isn't the industry freaking out about DeepSeek 4.1 Flash?
You can just use plenty of good handmade languages now, I'm not claiming mine will ever be good or useful to anyone other than myself. I'm using AI so heavily on my project because I want to explore the PL design space without spending years on implementation for things I'll probably want to throw away and rewrite a different way once I actually play around with them properly. And because I lack the motivation to persist through the sheer volume of grunt work that language implementation needs to get to the juicy interesting parts.
p1necone··on Why isn't the industry freaking out about DeepSeek 4.1 Flash?
I would probably go that route if I could use other harnesses with claude models, but I don't want to be locked in to claude code, and their API pricing (which you need to use it with other harnesses) is so much higher than subscription.
p1necone··on Why isn't the industry freaking out about DeepSeek 4.1 Flash?
I have a pretty large, complex project I've been building with heavy AI use (new language + compiler). I was following a 'strong model as orchestrator launching cheap models as implementers' pattern, but I recently trialled just using Deepseek-V4.1-Flash as the model for both layers because of the cost savings (with mimo v2.6 flash on code review agents for some decorrelation).

I was previously using GLM-5.3 as the orchestrator, after switching to DS anecdotally there was an unnacceptable quality loss, mostly around not taking all the relevant context into account when making decisions, pulling new design out of thin air without discussion too often, and being way too wordy and rambly in documentation despite prompting to avoid it. There's a lot of docs, rulings, core concepts, design philosophy to uphold and DS was just not cutting it.

However, it's perfectly capable of being the sole agent for all of my well specced implementation tasks. I've gone back to GLM as the orchestrator.

p1necone··on Meta and Microsoft take steps to reduce employee usage of Claude AI
Those things fall under the 'best in class' calculus for me.
p1necone··on Meta and Microsoft take steps to reduce employee usage of Claude AI
Only if all you care about is developing models. I assume the rest of the business would rather just use whatever's best in class regardless of who made it, so I'm sure it's not that straightforward of a decision.
p1necone··on The Future of Mathematics
It's the 'can answer arbitrary questions' part for me.

You could give me the most well crafted, comprehensive document ever and I would still misread bits of it, or miss some crucial bit of information that contextualises the rest or whatever.

With LLMs I can ask endless dumb questions in a way that just would not be practical otherwise. The quality of the responses doesn't even have to be particularly good for it to be extremely useful - it's basically a turbo charged rubber duck for learning new concepts.

p1necone··on Improper redaction reveals Google Data Center water and electricity usage
Depends on where it's coming from and who else needs water. If the water supply being drawn from was collected from rainwater initially and there's still enough for local residents to also have water, then sure, it's probably sustainable. But if the water is coming from an underground aquifer and there's otherwise not enough rain in the region to go around then it's a finite resource.
p1necone··on Show HN: Koi.rest – watch some fish and regain your balance
I noticed the koi pop in and out of existence on visible parts of the screen - it would be nice if incoming koi swam into frame from offscreen, and outgoing koi swam offscreen before disappearing.
p1necone··on Roboharm: Do frontier robot policies refuse unsafe instructions?
The baby one is pretty dumb, but the rest seem like decent tests, although a really smart model would probably realise this is some kind of staged test and not a real situation in all of them.
p1necone··on AI coding has made CI a bottleneck, so we reworked ours to keep up
If every software company saw a roughly equal improvement in productivity, they'd still be splitting the same customer base amongst each other - not clear whether revenue would actually go up for anyone.

Perhaps some new markets could be entered that weren't feasible before?

p1necone··on Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash
Given the title of the post says 'can match deepseek v4 flash' I think it's fair to call out these sort of dumb mistakes.
p1necone··on Training a 4B model to produce 81% faster query plans than Postgres
Having the moral high ground matters less than having a big warchest of money to spend on lawyers.
p1necone··on Stabilizing Rust's Never Type
Yeah I think just calling it 'Never' is much clearer, this isn't something that needs a dedicated single character, and ! is less readable imo
p1necone··on Everything I own, owned
Yeah OLEDs are the first display tech we've had that is on par or better than CRTs in all aspects. On balance the tradeoffs were definitely worth it for the various LCD panel options but motion clarity, colour intensity and black levels were all still beaten by a CRT in a dark room until we got OLEDs.
p1necone··on Everything I own, owned
If OLEDs are 'designed to fail' they've certainly gone about it in a very roundabout way - there would be much easier ways to do planned obsolescence if that was actually the goal of display manufacturers.
p1necone··on There's no such thing as a small software team anymore
Imo there's zero difference between 'thousands of microservices' and 'one monolith with thousands of functions/classes/etc' in terms of effort required to make it correct - both are still tiny interconnected pieces making up a whole, and you can separately unit test a function just as easily as you can separately e2e test a single microservice.

They still make up some whole product that presumably does something as a whole that you actually care about, and the complexity inherent to that doesn't go away with either approach.

p1necone··on OpenRouter is joining Stripe
It seems like a misunderstanding of basic economics to assume that the price for this would track FTE salaries.
p1necone··on DeepSeek V4 Pro 0813
This has not been my experience. Generally I do pin to 1 provider, or 1 provider with a couple fallbacks (especially with deepseek - most providers are 10x the cached token price compared to deepseek themselves), but even when I don't I still usually see 99%+ cache hit percentage. Specifically using pi with various ad-hoc customisations (that I was careful not to break prompt caching with).
p1necone··on DeepSeek V4 Pro 0813
It's interesting that all three of those used roughly the same amount of tokens, and almost entirely output. Feels like the thinking level lever didn't alter cost at all for this specific task, even though it did change the output.
p1necone··on DeepSeek V4 Pro 0813
50% cache hit is really low - in a standard agentic loop you should expect like 99%+ cache hit percentage (which should also lower that $12.50 to like a couple of $ for the same amount of tokens).

If you're using a customised harness you should make sure you don't have something that's e.g. changing your system prompt on some requests or rewriting history - it can be tempting to do stuff like strip old thinking tokens or compact tool call results to reduce context size but it's a trap - you want to never change history because of how cheap cache is, even more so with deepseek because their cache hit pricing is so low compared to most other models.

p1necone··on Show HN: iPhone app takes simultaneous images from 2 lenses, fuses into 1 photo
Apple customers are just built different.
p1necone··on Show HN: iPhone app takes simultaneous images from 2 lenses, fuses into 1 photo
I just assumed this was what all phones with multiple rear cameras were doing, is it not? What are the multiple cameras for other than that?
p1necone··on DeepSeek V4 Flash 0731
I read that and thought "ah that's going to confuse people, but I can tell they mean 30k-250k loc", so thank you for the clarification.
p1necone··on 2027 memory capacity is reportedly sold out
The capital investment needed to bring new chip fabs online and to staff them is likely orders of magnitude higher than that needed to buy land to grow corn on. And then the ratio of investment to sell price on that land + infrastructure is probably significantly worse for chip fabs that potentially aren't needed to satisfy demand anymore a few years from now.

"If the price is high enough" is of course technically true, but the scale of what high means in this context is important.

p1necone··on Pi's Minimalism Is Its Advantage
I love hacking away at pi extensions.

I realised how use case dependent harness behaviour is when I tried to use my customised-for-a-side-project pi config at work and realised I needed to tweak it significantly to be useful - I would not be surprised if tools like Claude Code needing to be all things for all people is hurting their peak usefulness.

p1necone··on Developers are attached to tools because tools encode trust
I'm speaking in the context of building scaffolding for e2e testing, not the tests themselves.

When I say "everything" I mean the frontend, backend, database, seeded data, optimizing runtime etc, I'm not talking about loc coverage. I would not recommend using llms to generate the tests themselves without heavy guidance and review, and you're right that 100% coverage is often a counterproductive goal if the code under test is anything less than pure business logic.

p1necone··on Show HN: Shitty – fast terminal. Memory-unsafe and faster than yours
This is cool, but I gotta say - I care much more about keypress-to-screen latency on my terminals than throughput - would love to see some numbers on that.
p1necone··on Developers are attached to tools because tools encode trust
I've found the process of building good reliable CI that thoroughly covers everything has been greatly improved by LLMs. There's so much tedious plumbing and grunt work involved in building CI and automated testing infrastructure for bespoke products that they can handle just fine while you concentrate on the important bits - I would say it's an area where agentic workflows are even more suited than regular product code.
← PreviousPage 2 of 34Next →