frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Tacto One – open-source Hand Exoskeleton

https://github.com/molanocortes/takto-one/tree/main
1•_Microft•51s ago•0 comments

Ask HN: What's it like working at Google nowadays? Core teams like Search, Ads?

1•karakoram•1m ago•0 comments

Bring your website or project to Lovable

https://lovable.dev/free-import
1•doener•2m ago•0 comments

What YC's Minecraft and Mindustry agent swarms show about multi-agent teams

https://theainewsreport.com/2026-10-30-yc-agent-swarms-minecraft-mindustry-explained.html
1•titusblair•3m ago•0 comments

Disk I/O profiling with eBPF

https://coroot.com/blog/disk-io-profiling-with-ebpf/
2•nikolay_sivko•4m ago•0 comments

Binance hot wallet signed 50 EIP-7702 delegations 3 minutes before the drain

https://www.reddit.com/r/CryptoCurrency/comments/1xj0mp0/binance_hot_wallet_signed_50_eip7702_del...
1•Onavo•5m ago•0 comments

Show HN: Teaching a 4M-param model to read like a child, with RL and no labels

https://rjb7731.github.io/blog/2026/learning-from-scratch/
1•rybthrow2•5m ago•0 comments

Check for IOCs for the ongoing Claude Code extensions exfil campaign

https://github.com/cookiengineer/check-claude-code-exfil
1•cookiengineer•5m ago•0 comments

Recent and Cumulative Cannabis Use and Sleep in Middle-Aged Adults

https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2855118
1•bookofjoe•8m ago•0 comments

More thinking isn't (always) better

https://www.ariwilson.com/writing/agent-effort-levels/
1•ariwilson•9m ago•1 comments

Signet: Cryptographic Enforcement of LLM Transitions

https://github.com/harishbalasubramanian/signet-ai
1•sterling-vance•11m ago•0 comments

What happens when cloud coding agents are as fast as your local machine?

https://opencomputer.dev/blog/cloud-agents-as-fast-as-local/
1•iacguy•11m ago•0 comments

The $200M Machine that Prints Microchips [video]

https://www.youtube.com/watch?v=B2482h_TNwg
1•tosh•12m ago•0 comments

U.S. Oil Industry Earned over $130B in Profit During Iran War

https://www.nytimes.com/2026/10/30/business/energy-environment/exxon-chevron-oil-profits-iran-war...
7•ceejayoz•13m ago•0 comments

First video gen model trained on real neurons online

https://twitter.com/AlexKsendzovsky/status/2116153410101944352
1•philippb•15m ago•0 comments

Show HN: I built a safety gate for autonomous coding agents

https://inbin-gate.vercel.app/
1•emmanueltsakpo•15m ago•0 comments

Show HN: Hashsync – Deduplicate SSH streams against the receiver's files

https://github.com/someplaceguy/hashsync
1•someplaceguy•16m ago•0 comments

OpenAI Private Intelligence

https://openai.com/business/private-intelligence/
1•pretext•19m ago•0 comments

A Working Nintendo in ~1GB of Qwen Weights

https://github.com/dbreunig/nesformer
8•dbreunig•21m ago•3 comments

Voyager 1 gives controllers the silent treatment

https://www.theregister.com/science/2026/10/30/voyager-1-gives-controllers-the-silent-treatment/5...
4•ohjeez•22m ago•0 comments

Show HN: NKTg AI – An LLM that decodes text before generating

https://nktg.org/
1•NKTgAI•23m ago•0 comments

Google Jules Turndown on December 31, 2026

https://jules.google/docs/changelog/2026-10-29
3•simonpure•23m ago•0 comments

A geography game where you guess cities based on compass arrows

https://visitwhale.com/city-angle/
1•whalegames•24m ago•0 comments

Package science into reusable, citable knowledge

https://www.aluna.ai
1•Bluestein•25m ago•0 comments

Product Hunt is swamped now

https://www.mahl.me/blog/product-hunt-is-swamped-now/
1•monneyboi•26m ago•0 comments

Interface Lineage – library of interface patterns and principles

https://interfacelineage.com/
1•ChrisArchitect•26m ago•0 comments

Show HN: Shell-pet – a hermit crab for your README that reacts to repo activity

https://shell-pet.vercel.app
1•amhtutt•26m ago•0 comments

OpenAI CEO Sam Altman Says You Having DRAM Instead of Data Centers Is "Terrible"

https://www.thegamer.com/openai-ceo-sam-altman-consumers-having-dram-terrible/
9•josephcsible•27m ago•3 comments

Postgres 19 is late, and it's AI's fault

http://rhaas.blogspot.com/2026/10/postgres-19-is-late-and-its-ais-fault.html
1•ingve•27m ago•0 comments

OpenRouter rebuilt provider routing for better, faster inference

https://openrouter.ai/blog/announcements/how-we-rebuilt-provider-routing/
1•tosh•29m ago•0 comments
Open in hackernews

US tech stocks see sharp falls over AI cost fears

https://www.bbc.co.uk/news/articles/cvwy72e0n72yo
21•mindracer•49m ago

Comments

elpakal•35m ago
Google's AI search is abysmal and is almost always, at least, 30% wrong (outdated, wrong info) ime. I find that I just use the chat interface to save tokens on basic things so I don't want to waste my real AI subscription tokens on basics.

In theory I understand their change to make search -> AI search instead, but I have a hard time believing that this was benchmarked in any way and not rushed to desperately remain relevant in the AI race.

eigenspace•27m ago
People say this a lot, and part of me wants it to be true and hates this AI search stuff being shoved in my face, but this really hasn't been my experience.

Unless I'm searching for something very niche, I'm often quite surprised by how 'directionally right' Google's AI overview is, even if I retain a lot of the skepticism I had towards it.

cush•12m ago
Other AI search products are so good. I’m not sure why Google is still so far behind. You can feel Gemini struggle in in the new Siri too, it’s awful
downrightmike•9m ago
They don't have anyone who fights for the consumer. They just make things and rename it all the time. No one knows what their thing is called today
foldr•12m ago
Like the sibling, I can't really agree with this. Google is now quite good at answering general queries and doing simple calculations, which is a big step up from its capabilities a few years ago. For example, you can ask

> What is the equivalent focal length on APS-C of 80mm on full frame?

and get the correct answer. It's not good compared to frontier AI models, but that's to be expected, given that you don't pay for it.

esafak•9m ago
It's also multi-modal, right?
Dennip•2m ago
Disagree. In my experience particularly over the last few weeks, Gemini has gotten very good at understanding what i'm asking for and giving a relevant response that 9 times out of 10 just answers my question without me having to go further into the relevant wiki link etc.
feverzsj•1m ago
It's mostly wrong, if you dig a little bit into the supplied references.