frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Schlep Blindness (2012)

https://www.paulgraham.com/schlep.html
1•ksec•2m ago•0 comments

Show HN: OSS app to find LLMs across multiple LLM providers (Azure, AWS, etc.)

https://github.com/mozilla-ai/any-llm/tree/main/demos/finder
2•njbrake•7m ago•0 comments

Show HN: I built a way to monetize any link with a crypto paywall

2•allynjalford•8m ago•0 comments

Geoffrey Hinton: 'AI will make a few people much richer and most people poorer'

https://www.ft.com/content/31feb335-4945-475e-baaa-3b880d9cf8ce
1•pseudolus•8m ago•1 comments

Orchestrate multiple AI agents with cagent by Docker to create coding assistant

https://tobiasfenster.io/orchestrate-multiple-ai-agents-with-cagent-by-docker
2•AsmodiusVI•17m ago•2 comments

Windows-Use: an AI agent that interacts with Windows at GUI layer

https://github.com/CursorTouch/Windows-Use
2•djhu9•21m ago•0 comments

Anthropic is endorsing SB 53

https://www.anthropic.com/news/anthropic-is-endorsing-sb-53
1•arroia•21m ago•0 comments

But how do AI images and videos work?

https://www.youtube.com/watch?v=iv-5mZ_9CPY
1•tzury•24m ago•0 comments

Biz Academy aiuta imprenditrici con percorsi online

https://biz-academy.it/i-top-20-migliori-podcast-italiani-da-ascoltare-subito/
1•lorisfreez•24m ago•0 comments

Internet censorship is complex, usually

https://en.wikipedia.org/wiki/Internet_censorship
1•DaveZale•30m ago•1 comments

How Python Implements List Comprehensions

https://pythonkoans.substack.com/p/koan-11-the-flowing-river-part-2
2•meander_water•36m ago•0 comments

Jakub and Szymon

https://blog.samaltman.com/jakub-and-szymon
3•davidbarker•42m ago•0 comments

How Big Was IBM?

https://thechipletter.substack.com/p/how-big-was-ibm
3•chmaynard•44m ago•1 comments

It's AI all the way down as Google's AI cites web pages written by AI

https://www.theregister.com/2025/09/07/googles_ai_cites_written_by_ai/
1•akyuu•45m ago•0 comments

Agentic AI Runs on Tools

https://simplicityissota.substack.com/p/agentic-ai-runs-on-tools
1•bookish•45m ago•0 comments

XFS File-System Ready to Enable Online Fsck Support by Default

https://www.phoronix.com/news/XFS-Ready-Online-FSCK-Default
2•Bender•45m ago•0 comments

New Linux Patches Enhance Intel Nested Virtualization Performance on Linux

https://www.phoronix.com/news/Linux-KVM-Nested-Intel-VMX-Perf
2•Bender•46m ago•0 comments

Ask HN: Where can I learn about T-Splines?

1•dschroer•46m ago•0 comments

Scaling the system that powers over $700M ARR invoices

https://medium.com/@chandnij/scaling-the-system-that-powers-over-700m-arr-invoices-596ac47a6983
1•openmosix•46m ago•0 comments

Fatal stabbing of Ukrainian refugee in North Carolina ignites crime debate

https://www.bbc.com/news/articles/cgknxyl77x6o
5•Bender•50m ago•0 comments

The Potato Chip That Destroyed the Bowels of America

https://www.cracked.com/article_28476_the-potato-chip-that-destroyed-bowels-america.html
3•masterjack•52m ago•0 comments

Is AI's Next Big Leap, Agentic AI, Set to Revolutionize or Ruin Tech Innovation?

https://roundtable.now/chats/84cc5f4e-84ba-4c74-a57c-6058f9218c63
1•soh3il•1h ago•0 comments

Kubernetes Primer: Dynamic Resource Allocation (DRA) for GPU Workloads

https://thenewstack.io/kubernetes-primer-dynamic-resource-allocation-dra-for-gpu-workloads/
2•nkko•1h ago•0 comments

Show HN: Astrae – Landing Page Templates for React, Tailwind, and Framer Motion

https://www.astrae.design
1•aretecodes•1h ago•0 comments

The Storm Hits the Art Market

https://news.artnet.com/market/intelligence-report-storm-2025-2684512
32•onecommentman•1h ago•14 comments

Does the Car Wash I Bought 1 Year Ago Make Money?

https://www.youtube.com/watch?v=FbW9rhVzRsU
1•bane•1h ago•0 comments

Goldman Sachs Warns of 1.9M Barrel per Day Oil Glut by 2026

https://oilprice.com/Energy/Energy-General/Goldman-Sachs-Warns-of-19-Million-Bpd-Oil-Glut-by-2026...
2•toomuchtodo•1h ago•1 comments

Tesla market share in US drops to lowest since 2017

https://www.reuters.com/business/autos-transportation/tesla-market-share-us-drops-lowest-since-20...
43•nabla9•1h ago•30 comments

Show HN: I Built Logstalgia for the Web

https://tailstream.io
1•arondeparon•1h ago•0 comments

Photoshop Is Dead Long Live Photoshop

https://www.mikechambers.com/blog/post/2025-09-07-photoshop-is-dead-long-live-photoshop/
1•mesh•1h ago•0 comments
Open in hackernews

Microsoft doubles down on small modular reactors and fusion energy

https://www.techradar.com/pro/microsoft-joins-world-nuclear-association-as-it-doubles-down-on-small-modular-reactors-and-fusion-energy
39•mikece•5h ago

Comments

rhdhfjej•4h ago
Remember when they told us in CS class that it's better to design more efficient algorithms than to buy a faster CPU? Well here we are building nuclear reactors to run our brute force "scaled" LLMs. Really, really dumb.
DeepYogurt•3h ago
Big O? More like Big H2O (heavy water).... I'll see myself out.
JumpCrisscross•3h ago
> Remember when they told us in CS class that it's better to design more efficient algorithms than to buy a faster CPU?

No? The tradeoff is entirely one between the value of labour versus the value of industry. If dev hours are cheap and CPUs expensive. If it’s the other way, which it is in AI, you buy more CPUs and GPUs.

estimator7292•3h ago
This makes sense if and only if you entirely ignore all secondary and tertiary effects of your choices.

Things like massively increased energy cost, strain on the grid, depriving local citizens of resources for your datacenter, and let's not forget ewaste, pollution from higher energy use, pollution caused by manufacturing more and more chips, pollution and cost of shipping more and more chips across the planet.

Yeah, it's so cheap as to be nearly free.

codingrightnow•3h ago
The massive amount of autists on the website leads to a lot of the humanity and emotion behind discusion being completely missing. Nearly everything comes down to an analysis of data and practicality.
JumpCrisscross•3h ago
> it's so cheap as to be nearly free

Both chips and developer time are expensive. Massively so, both in direct cost and secondary and tertiary elements. (If you think hiring more developers to optimise code has no knock-on effects, I have a bridge to sell you.)

There isn't an iron law about developer time being less valuable than chips. When chip progress stagnates, we tend towards optimising. When the developer pipeline is constrained, e.g. when a new frontier opens, we tend towards favouring exploration over optimisation.

If a CS programme is teaching someone to always try to optimise an algorithm versus consider whether hardware might be the limitation, it's not a very good one. In this case, when it comes to AI, there is massive investment going into trying to find more efficient training and inference algorithms. Research which, ironically enough, generally requires access to energy.

yannyu•2h ago
> Things like massively increased energy cost, strain on the grid

This is a peculiarly USA-localized problem. For a large number of reasons, datacenters are going up all over the world now, and proportionally more of them are outside the US than has been the case historically. And a lot of these places have easier access to cheaper, cleaner power with modernized grids capable of handling it.

> pollution from higher energy use

Somewhat coincidentally as well, energy costs in China and the EU are projected to go down significantly over the the next 10 years due to solar and renewables, where it's not so clear that's going to happen in the US.

As for the rest of the arguments around chip manufacturing and shipping and everything else, well, what do you expect? That we would just stop making chips? We only stopped using horses for transportation when we invented cars. I don't yet see what's going to replace our need for computing.

utyop22•55m ago
"Which it is in AI, you buy more CPUs and GPUs."

Ermmm. what?

infecto•3h ago
Pretty exciting to me. Constraints breed innovation and it’s possible that the wave of AI leads to new breakthroughs on the green energy front.

Edit: Amazing how anti-innovation and science folks are on HN.

logicchains•3h ago
There's no efficient algorithm for simulating a human brain, and you certainly haven't invented one so you've got absolutely no excuse to act smug about it. LLMs are already within an order of magnitude of the energy efficiency as the human brain, it's probably not possible to make them much more efficient algorithmically.
irjustin•3h ago
I'm really sad the core argument for the Matrix's existence doesn't hold up (it never did, just for me as a kid is all).
juliangamble•3h ago
They started with a different, more brilliant idea, of using human brains as a giant neural net, then backed away from that: https://news.ycombinator.com/item?id=12508832
irjustin•1h ago
oh wow that's cool. I do understand why they moved away from it. Battery is waaaayyyyy easier to understand for the layman and lay-kid (me).
rhdhfjej•3h ago
Your brain has a TDP of 15W while frontier LLMs require on the order of megawatts. That's 5-6 orders of magnitude difference, despite our semiconductors having a lithographic feature size that's also orders of magnitude smaller than our biological neurons. You should do some more research.
adrian_b•3h ago
The TDP of a typical human brain is not 15 W, but 25 W, so about the same as for many notebook or mini-PC CPUs, but otherwise your argument stands.

The idle power consumption of a human is around 100 W.

crinkly•3h ago
The one company I really want to see involved in dangerous things with clean up and serious environmental risks is the company that has serious production QA problems, an attention span of about 2 minutes, regular bouts of corporate schizophrenia and a policy of forcing half the planet to abandon working hardware.

Nothing good can come of this.

Microsoft needs to start asking if it should do something before it does it.

nomel•3h ago
Or, for another perspective, they're helping force capitalism to work: an opportunity exists to reduce prices.

This is a good way to force the (often monopolistic) providers to get their shit together, as google did with google fiber.

crinkly•3h ago
Ah yes a nuclear race to the bottom. Sorry but that will be a monumental fuck up if you have any historical knowledge about how we handle nuclear materials.

Capitalism fails very quickly the moment you try and push past sensible regulation and legislation. Look at the whole US situation right now.

It's expensive as hell already and we still don't handle waste or environmental issues properly. Capitalism isn't going to solve anything other than the price as it'll defer the rest until it's someone else's problem much like it does not on every single damn sector's waste.

I'm not anti-nuclear. We need it. What we don't need is tech companies getting into the market.

pfdietz•3h ago
We do not in fact need it.
nomel•2h ago
Microsoft, Google, and Amazon are interested in it enough to put money into it. That suggests there is a market need for it, or at least interest, within the current context.

We should assume they're acting rationally, so the real question is, why do they find this interesting at all? Why not dump the money into private solar farms instead?

pfdietz•4m ago
The world will spend about a quadrillion dollars on energy in the 21st century. Piddling little billion dollar investments look like long shot bets on low probability outcomes. The big spending is on renewables now.

Gates in particular seems to have been a disciple of Vaclav Smil, a person whose arguments against renewable cost reduction were wildly mistaken.

RandallBrown•3h ago
Are there serious environment risks with fusion reactors?

My understanding was that very little radioactive waste was created from a fusion reactor and what little there is will decay pretty quickly (decades).

crinkly•3h ago
Well we don't have working fusion reactor topology yet but the current "reactor" components are low level waste so safe within 40-100 years. Which is still a hell of a long time. Also they still will require biological shielding and associated materials are quite difficult to deal with (concrete etc).

I expect that the longevity of their attention is considerably less than this, particularly if the LLM boom crashes. ROI will not pay for the disposal later down the line.

pfdietz•3h ago
The claims of low level waste from fusion reactors implicitly assume that impurity elements (like, say, niobium) that would produce long lived activation products can be reduced to very low levels. This may drive up the cost of materials dramatically.
philipkglass•3h ago
You are correct. Radioactive materials from fusion reactors are not a significant environmental threat. The bigger problem with fusion reactors is that nobody has yet built a controlled terrestrial fusion reactor that produces net power.
pfdietz•3h ago
The big risk is tritium leakage.

To show the scale of the problem: if the world were powered by Helion's reactors (for all primary energy), and the tritium produced were just released into the environment and mixed completely with all water on the planet (including oceans, lakes, rivers, ground water, and ice), then it would lift all that water above the US regulatory limit for tritium in drinking water. All the water, including everything in every ocean.

adrian_b•2h ago
A fusion reactor creates a much more intense flux of neutrons than any fission reactor, which will transmute into radioactive isotopes any substance from which a shield will be made.

So the quantity of radioactive waste will certainly not be little, but more likely much greater than in a fission reactor.

Nevertheless, because there is more freedom in the design of the neutron shield than in a fission reactor, it is likely that it is possible to find such compositions where most of the radioactive waste will decay quickly enough, so that there will remain only a small quantity of long-lived radioactive waste.

However, until someone demonstrates this in reality, it is still uncertain how much radioactive waste will be generated, because this depends on many constructive details.

A lot of components of a fusion reactor, e.g. pipes for cooling fluid and the like, will become damaged by the neutrons and they will have to be replaced periodically, after becoming radioactive. The amount of such waste will depend a lot on the lifetimes of such components. For now it is very uncertain how much time such components will resist before requiring maintenance.

utyop22•1h ago
"Microsoft needs to start asking if it should do something before it does it."

Do they? I hope they don't. I would enjoy seeing MSFT implode and losing trust of its shareholders with its cash - itll be forced to return it rather than reinvest.

hereme888•3h ago
Yay, common-sense energy ftw. Good for MS.
mclau157•3h ago
But fusion has never been proven to work at scale
messe•3h ago
On the contrary, I think it has only been proven to work at scale: https://en.m.wikipedia.org/wiki/Star
cladopa•3h ago
It has been:

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

tootie•3h ago
From what I can tell, Helion energy has already broken ground on what would be a commercial fusion reactor connected to the WA grid despite their best prototype still not producing net positive energy. It's a gamble, but presumably everyone involved is willing to take the risks. A data center that runs on fusion would be a real watershed moment and everyone wants to be first.
philipkglass•3h ago
If Helion delivers a working fusion reactor that produces net electricity, at commercially competitive rates, I think that's an even more significant event than the recent AI boom.
tim333•2h ago
Helion is a odd one. They have picture of the site here https://www.helionenergy.com/articles/helion-secures-land-an...

saying "Milestone keeps Helion on track to deliver electricity from fusion to Microsoft by 2028"

but as you say they don't seem to have produced any energy and after watching Sabine's take I'm very skeptical (https://youtu.be/YxuPkDOuiM4)

I think it may be a bit of a scam where they keep the investment and their jobs going as long as possible but don't produce power.

lazide•3h ago
Fusion definitely works at large scale; and in short bursts. That is the Sun and Thermonuclear weapons.

What has not yet been shown (and may be impossible?) is fusion working at small scale and over long timeframes.

winterismute•3h ago
I read this analysis of the SMR farm announcement in Canada a few months ago and I found it quite insightful: https://www.carboncommentary.com/blog/2025/5/11/the-first-te...
pfdietz•3h ago
"there is no evidence today that SMRs will reduce electricity costs compared to continuing rapid investment in wind and solar."
mikestorrent•3h ago
Not with the approach we are showing, but if solar was built like this, it would fail too: remember Solyndra? Treating it as a bespoke construction project instead of as a commodity manufacturing project is the fundamental mistake that continues to result in nuclear costing too much.

Fuck's sake, it's just some hot rocks boiling a kettle, we make it out to sound like it's magic but we had the technology for this ~80 years ago. By now we should have the cost of a standard issue nuclear plant down to way cheaper than anything else. Common layout, protocols, processes, software at all of them... could have been complete in 1989, honestly.

pfdietz•2h ago
But solar isn't built like nuclear. Solar involves parallel exploration of device designs at very small scale, installed with massive redundancy and resilience. Many billions of PV cells have been manufactured. The real cost decline driver is manufacturing automation. Nuclear, even SMRs, have orders of magnitude coarser granularity.

If you want "hot rocks", it's probably much cheaper to just resistively heat them with cheap solar (you don't even need inverters). This could store energy over many months and, pushed to its cost reduction limits this promises to be the final nail in the coffin for any dreams of a nuclear revival.

https://news.ycombinator.com/item?id=45012942

mikestorrent•14m ago
> Solar involves parallel exploration of device designs at very small scale, installed with massive redundancy and resilience.

I am imagining a field of shipping-container sized units, each of which is a small modular reactor. Probably with solar panels on top ;) Still a few orders of magnitude different, but the idea here is that each unit is small enough that it can be manufactured, so that nuclear plant bring-ups don't take 30 years. Most of the cost is because of the tremendous generational effort involved in just a single project; what does it take to reduce the cost of the plants themselves to the point where they can really shine, economically?

The goal is to have reliable base load power generation so that we don't have to deal with the massive complexity and carbon footprint of battery plants all over the place to deal with peaky generation technologies like solar. I don't believe that that is a solved problem: using tremendous amounts of rare earth materials for limited-lifespan installations that don't even produce energy is possibly not the best use of our resources, considering it's almost all fossil fuel going into those logistics operations anyway, right? EROEI for a battery plant is going to be hard to achieve.

mikestorrent•3h ago
Depressing, but it shows the typical faults of most Canadian projects these days. Massive government spend on a project doomed to fail by economic analysis before it's even online; and no takeaways for the Canadian people to actually get momentum going.

If we wanted to do SMRs right, the goal should be to build one or more SMR production factories, here in Canada, where we manufacture N reactors per month, that fit onto train cars, and can be delivered to qualified, secure sites around the world. Instead, we're paying massive cash out to GE Hitachi, and so the end result will never be "the capability of building and deploying SMRs", it will be "4 unprofitable SMRs in a facility and $4.4 billion a unit if we want more of them to lose money on".

Obviously this is doomed to fail; the units should cost like $100M max so they have positive ROI within a few years. If the unit will never beat solar in $/megawatt for operating and fueling costs, and won't pay for its own construction cost before its lifetime ends, it should never have been constructed; the entire thing is catabolic, all of the work and carbon that goes into it is an utter waste. Everyone involved should just do something else with their lives if we're going to approach it this way.

What's the point? Why do such small-minded people get authority over grand projects?

tomComb•2h ago
It’s usually about well connected companies lobbying for free money. It’s the sort of thing that keeps Bell and others afloat and guarantees they never have to get competitive.

The gross thing is seeing the public cheer it on.

mikestorrent•13m ago
I'm still half cheering it because at the very least it's still nuclear progress, and will help ensure we still have nuclear energy workers for another generation here. I worry a lot about what's been done to the Atomic Energy Workers in terms of whittling away at our capability to produce good energy workers with tribal wisdom and the Canadian nuclear culture of safety.
bryanlarsen•3h ago
Have we seen Microsoft actually put any skin in the game yet? All the pre-purchase announcements are virtually risk free for Microsoft. They've agreed to buy a certain amount of power at a certain price, if the counter-party can deliver it. But they're not pre-paying, they only pay when the electricity gets delivered. If they never deliver, Microsoft isn't out any money.
lazide•3h ago
It’s a smart move on their part. It’s also a way for VC/investors to have some concrete value prop in their math. Aka if x works, I’d get at least y return (where y is guaranteed to not be zero)
JumpCrisscross•2h ago
I'm halfway certain it's for political optics.

I'm intensely pro nuclear. But the tech is still in the stables. We need research into driving down costs. In the meantime, we need to think harder about where we're putting datacentres and how we can, if not make power cheaper for average Americans, at least not raise its real cost.

jmyeet•3h ago
Ugh, I rdread this topic because nuclear is as close to as a religion as you get on HN. SMRs just arne't better in any way that matters [1].

And while I personally hope we have economical commercial power generation in the future, I'm not convinced that'll ever happen due to one massive problem: energy loss from high-energy neutrons, which have the added problem that they destroy your very expensive containment vessel. Stars deal with this by being massive, having fusion happen in the core (depending on the size of the star) and gravity, none of which is applicable to a fusion reactor.

I'm reminded of the push recycling of plastic. Evidence has surface that this was nothing more than oil industry propaganda to sell more plastic [2]. A lot of "recycling" is simply dumping the problem into developing countries and then just looking the other way. We used to do this to China until they stopped taking plastic to "recycle".

I can't help but think that Microsoft issuing some press releases about nuclear is nothing more than marketing to contributing to the data center explosion that will inevitably drive up your electricity bills because you'll pay for the infrastructure that needs to be built and will be paying the generous (and usually secret) subsidies these data centers engotiate.

[1]: https://blog.ucs.org/edwin-lyman/five-things-the-nuclear-bro...

[2]: https://www.npr.org/2020/09/11/897692090/how-big-oil-misled-...

ChrisArchitect•35m ago
Announcement release: https://world-nuclear.org/news-and-media/press-statements/wo...