frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Show HN: audio.observer – AI-generated songs from current events

https://audio.observer/
1•ugnju•59s ago•0 comments

Show HN: Symbols Copy and Paste

https://copyasymbol.com/
1•artiomyak•2m ago•0 comments

Benchmarking Subquadratic's latest model and SSA Kernel

https://www.appen.com/whitepapers/benchmarking-subquadratics-latest-model-ssa-kernel
1•famouswaffles•2m ago•0 comments

PoC: VPN over WebRTC to Bypass Whitelists

https://github.com/kulikov0/whitelist-bypass,https://github.com/kulikov0/whitelist-bypass-iran
2•kulikov0•3m ago•2 comments

Show HN: I built an open-source tool to see if ChatGPT recommends your products

https://github.com/foodaka/openllmrank
1•zipvoila•3m ago•0 comments

Understanding the Modern Cybercrime Landscape

https://www.technologyreview.com/2026/05/19/1136925/understanding-the-modern-cybercrime-landscape/
1•joozio•5m ago•0 comments

Show HN: LibreOffice-rs – I built a pure-Rust LibreOffice using autoresearch

https://github.com/clark-labs-inc/libreoffice-rs
2•stan_kirdey•6m ago•0 comments

The AI economy is rewriting the American Dream; and blue-collar is poised to win

https://www.cnbc.com/2026/05/19/ai-hiring-slowdown-skilled-trade-workers.html
1•indigodaddy•6m ago•0 comments

Google I/O

https://io.google/2026/
1•thanhhaimai•6m ago•0 comments

The New Workspace: A First-Principle Exploration of Dictation, Agents and Humans

https://www.inferterra.com/the-new-workspace-a-first-principles-exploration-of-dictation-agents-a...
1•matt_teresi•6m ago•1 comments

Confuse some SSH bots and make botters block you

https://ai.realhackers.org/confuse-some-ssh-bots.html
1•Bender•6m ago•0 comments

Raindrop Workshop: Your local OSS agent debugger

https://github.com/raindrop-ai/workshop
3•jamest•6m ago•0 comments

Show HN: Ait – Claude, Codex, and Aider as a team, on your laptop

https://github.com/m24927605/ait
1•m24927605•8m ago•1 comments

Google Flow Beta for Android

https://play.google.com/store/apps/details?id=com.google.android.apps.labs.whisk&hl=en_US
2•julianpye•11m ago•0 comments

A 1955 Los Alamos computer experiment changed our understanding of chaos

https://www.lanl.gov/media/publications/1663/science-of-unpredictability
1•LAsteNERD•12m ago•1 comments

Google IO 26 Keynote

https://www.youtube.com/watch?v=wYSncx9zLIU
3•Dinux•14m ago•0 comments

Polymarket debuts prediction markets tied to private companies

https://www.reuters.com/legal/government/polymarket-debuts-prediction-markets-tied-private-compan...
1•thm•15m ago•0 comments

Political Money Is Flowing to Influencers. But from Whom?

https://www.nytimes.com/2026/05/16/business/media/influencers-political-financing-disclosure.html
1•thm•16m ago•1 comments

Drone Swarms: Uncomfortably Plausible

https://silencingmachine.org/posts/0004-its-not-the-terminator-its-worse/
1•jaybill•16m ago•0 comments

Ask HN: Reliability Issues with AWS

2•yonisto•16m ago•0 comments

We made our sandbox filesystem 47× faster by deleting it

https://microsandbox.dev/blog/block-backed-rootfs
3•makeboss•16m ago•0 comments

Vendergood: Reconstructing a 1905 constructed language for AI agent cognition

https://github.com/mekickdemons-creator/libro-vendergood
1•Mekickdemons•17m ago•1 comments

Together, Edera and Minimus Claim They Can Protect Your Software from AI Hackers

https://cloudnativenow.com/features/together-edera-and-minimus-claim-they-can-protect-your-softwa...
1•CrankyBear•17m ago•0 comments

What's Easy Now? What's Hard Now?

https://brooker.co.za/blog/2026/05/18/whats-easy-whats-hard.html
2•KraftyOne•17m ago•0 comments

Bill C-22 Surveils Ordinary Canadians While Leaving Cartel Networks Untouched

https://www.thebureau.news/p/bill-c-22-surveils-ordinary-canadians
1•laurex•18m ago•0 comments

xs

https://cryptm.org/xs/
1•tosh•19m ago•0 comments

Anyone else been seeing any Google branding changes today?

1•dragonsenseiguy•20m ago•0 comments

Show HN: Agent threads – Share Claude Code and Codex sessions as public links

https://agent-thread.com
1•pixxxel•20m ago•0 comments

Dynamic Filters: 25x Faster Queries by Passing Info Between Operators

https://datafusion.apache.org/blog/2025/09/10/dynamic-filters/
1•killme2008•21m ago•0 comments

Trump Mobile Phone Is Finally Released–With a Major Blunder

https://www.thedailybeast.com/trump-mobile-phone-is-finally-releasedwith-a-major-blunder/
2•cf100clunk•21m ago•1 comments
Open in hackernews

Show HN: Heard Google copied part of our product for IO. Want to show off first

4•echelon•39m ago
I just heard from a very reputable source that Google built part of our product and that they're about to show it off at Google IO.

We built a 3D film tool called ArtCraft:

https://getartcraft.com

https://github.com/storytold/artcraft (our monorepo with the desktop app, server, and website code)

Our 3D virtual film set enables set decoration, location reuse, precision blocking (control over where things are), character posing, etc.

Sets and objects can be generated as meshes or gaussian splats.

Autoregressive multimodal models can treat input "previz" style scenes as ControlNets of sorts. Combined with reference images and a possibly a few follow-up rounds of editing, you can create extremely precise starting frames that reflect your vision as a director or creative almost exactly as you see it.

It gets rid of the "roulette wheel" of prompting in the places and times where you know you what you want. It's like a WYSIWYG editor.

Here's an example:

https://app.getartcraft.com/edit-3d/m_qa72baw3crghyfn2bbw52j... (Sci-Fi horror film, starring Garry Tan.)

I was just told that Google built a 3D tool similar to ours that they're showing off at Google IO. I don't view this as bad, but I do want to share ours before they get all the attention.

OpenAI's GPT Image is a lot better at this "3D previz to full render" task than Google's Nano Banana, and it's excelled at this workflow since GPT Image 1. (Unless Google shows off a brand new model for the task, I expect this to still be the case.)

Our plan is to build vertical specialized tools for animation, editing, dubbing and make the entire stack - including cloud API routers and model containers - open source.

Here's an example of timeline based animation (and we have the ability to edit animation curves for all objects, meshes, cameras, etc.):

https://imgur.com/a/00q76j2

Our tool has BYOK and subscriptions, so you can use lots of other API providers and even log in with the commercial versions of other models and aggregators. We don't care if we book the revenue directly, we want to build the rails and the UX and win mindshare.

We hit a pretty good growth spurt and have had folks start sending us PRs, which has been exciting.

Eager to see what Google's version of this looks like.

If you're a Rust developer interested in film and media, I'd love to say hi.