HackerLangs
TopNewTrendsCommentsPastAskShowJobs

nektro

655 karmajoined 8 tahun yang lalu
github.com/nektro

comments

nektro
·5 hari yang lalu·discuss
url should be https://ui.shadcn.com/docs/changelog/2026-07-base-ui-default
nektro
·10 hari yang lalu·discuss
saw this title on the front page and thought it meant the element, not a company
nektro
·10 hari yang lalu·discuss
one of the most ethically barren products to come out of this era
nektro
·16 hari yang lalu·discuss
thank you!
nektro
·28 hari yang lalu·discuss
the public only wins once we shut it down globally through treaties like other tech that's too dangerous for anyone to have
nektro
·bulan lalu·discuss
haven't written up a show hn post yet but i launched this weekend a server app for git i built all in Zig[1]

[1]: https://astheno.software/shale/

demo site https://git.mirrors.nektro.net/
nektro
·bulan lalu·discuss
> but do people realize that code is just a means to an end?

yes and an end has to be worth the means. people disagree about whether it is worth it. some people see a benefit using LLMs and agents to such a degree they're willing to look past the noise and water and air pollution produced by the unprecedented data center buildout that we've seen over the past few years, or the unprecedented increase in things like spam and deepfakes, or the way that it is decimating the job market + funnel, or the way that it reinforces existing antisocial incentives in society. some don't/aren't.
nektro
·bulan lalu·discuss
sota models cannot remotely fit in 128gb
nektro
·bulan lalu·discuss
anthropic rumored to be having a profitable quarter is a ruse to ready them up for IPO or otherwise. they could've had a profitable quarter at any point in their history but they only do so if they stop making models; this has been established. they make a model, they sell inference and make a ton of money, they train the next model at 10x the cost, they sell inference and make a ton of money, they train the next model at 10x the cost, repeat
nektro
·bulan lalu·discuss
> I would still recommend not putting all your eggs in one basket just yet because [..] there will still be some value in knowing how systems work, both to differentiate yourself from other developers career-wise, and as part of effective LLM steering.

the thesis is that investing in your skills outside of LLMs pays dividends whether you decide to apply those skills to LLMs or not, plus spending time bonding with your fellow engineers is good for you too. so I'm sure Zig will be doing great in a few years
nektro
·2 bulan yang lalu·discuss
not one mention of Zig on the whole page?
nektro
·2 bulan yang lalu·discuss
Alphafold isn't generative and using this as a rebuttal to OP is bad faith
nektro
·2 bulan yang lalu·discuss
so disgusting
nektro
·2 bulan yang lalu·discuss
woah, the web map viewer even has parallax when zoomed in
nektro
·2 bulan yang lalu·discuss
because open source is a means not an end. folks want good, fast software that respects them. open source isn't the only way there.
nektro
·2 bulan yang lalu·discuss
easy, CTFs should ban it. then it'd be more like the chess community
nektro
·2 bulan yang lalu·discuss
TIL <menu>, I wonder why more frameworks don't make use of it.
nektro
·2 bulan yang lalu·discuss
really enjoyed this talk. really interesting how the different constraints change how they think about their system
nektro
·2 bulan yang lalu·discuss
going after systems that affect students is beyond bad taste
nektro
·2 bulan yang lalu·discuss
dystopian :( i hope schools put more pressure on keeping that off their devices. or switch to neos.