frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Cezar: A parallel coding agents orchestrator

https://github.com/open-mercato/cezar
1•zaiste•1m ago•0 comments

AI Agents Governance: Who guards the guardrails?

https://www.nofire.ai/blog/who-audits-the-ai-agent
3•spirosoik•3m ago•0 comments

Senate Panel Has Obtained Copy of Fauci's Cellphone

https://www.nytimes.com/2026/08/05/us/politics/fauci-phone-senate.html
1•ianeff•3m ago•0 comments

A year of AI disclosure in critical packages

https://nesbitt.io/2026/08/06/a-year-of-ai-disclosure-in-critical-packages.html
1•omkar-foss•4m ago•0 comments

My Precious Attention

http://replicated.live/blog/click.html
1•gritzko•7m ago•0 comments

AI music generator Suno loses copyright infringement legal case

https://www.nme.com/news/music/ai-music-generator-suno-loses-copyright-infringement-legal-case-39...
1•passwordoops•8m ago•0 comments

Vista: A Visual Harness for Reasoning in an Interactive World

https://vista-research.github.io/
1•haxfenx•10m ago•0 comments

Cloudflare OS: great insight into how to embrace internal AI demand securely

https://cephalosec.com/blog/cloudflare-os-insights-on-embracing-internal-ai-growth-securely/
1•Versipelle•11m ago•0 comments

Time Magazine has a separate version of its website with ads only AI can see

https://www.theregister.com/ai-and-ml/2026/08/05/time-magazine-has-a-separate-version-of-its-webs...
1•rguiscard•11m ago•0 comments

Slow Light Communications

https://medium.com/luminasticity/slow-light-communications-81bdd7a87afe
1•bryanrasmussen•12m ago•0 comments

You Should Almost Never Use AI to Write Anything Substantive

https://www.erichgrunewald.com/posts/why-i-think-you-should-almost-never-use-ai-to-write-anything...
1•erwald•14m ago•0 comments

Stillsane: Detect silent quality drift in deployed LLM apps

https://pypi.org/project/stillsane/
1•MSanket9•15m ago•0 comments

Grounded document extraction with bounding-box citations

https://pspdfkit.github.io/nutrient-extraction-samples/
1•marakiii•16m ago•0 comments

Turning Air into Alcohol [video]

https://www.youtube.com/watch?v=ID-K64Jk6JM
1•DiscourseFan•17m ago•0 comments

HCC – local-first, cross-agent memory layer (98% cache hit, $40/mo)

https://github.com/icemaple77/hanyan-cognitive-core
1•icemaple77•18m ago•0 comments

Products make going back unimaginable

https://datawarrior.medium.com/the-best-products-make-going-back-unimaginable-30049de81568
1•nikolamilosevic•20m ago•0 comments

Chrysalis (hypothetical moon) and Saturn's rings

https://en.wikipedia.org/wiki/Chrysalis_(hypothetical_moon)
1•simonebrunozzi•21m ago•0 comments

Show HN: My free web dev courses, no signup, with exercises

https://en.andros.dev/courses/
1•andros•23m ago•0 comments

The Aikido Machine: on-prem AI pentesting that never leaves your network

https://www.aikido.dev/blog/the-aikido-machine-on-prem-ai-pentesting-that-never-leaves-your-network
1•ultra2d•24m ago•0 comments

Show HN: Rask – The .NET One Person Framework

https://pal-tamas.github.io/rask/
1•pal-tamas•26m ago•0 comments

Using Git diffs to circumvent Fable's safeguards

https://blog.j11y.io/2026-07-18_Jailbreaking-Fable-disinformation/
2•padolsey•27m ago•0 comments

ESPBanking – Offline, local banking server running on a single ESP32

https://github.com/dumasha05/ESPBBanking1/blob/main/README.md
1•Dumasha05•27m ago•0 comments

Anthropic is hiring an AI chip design team

https://techcrunch.com/2026/08/05/anthropic-is-hiring-an-ai-chip-design-team/
2•mpfect•27m ago•0 comments

Claude Fable 5 finds a tiny formula that topples an 87-year-old math conjecture

https://www.sciencedaily.com/releases/2026/08/260804034634.htm
3•theanonymousone•31m ago•0 comments

1 RCE, 2 AI models, 0 ways to tell who was right

https://zhenyi.gibber.blog/1-rce-2-ai-models-0-ways-to-tell-who-was-right
1•zhenyi•31m ago•0 comments

We built a local SQL engine for DynamoDB

https://dynotable.com/blog/local-sql-engine-for-dynamodb
2•dynotable•31m ago•1 comments

Train the AI to Want to Die

https://slimemoldtimemold.com/2026/08/05/a-stupid-idea-for-ai-alignment-we-came-up-with-by-lookin...
1•mrcgnc•34m ago•0 comments

Fable 5 on ARC-AGI (Verified)

https://twitter.com/arcprize/status/2085115252035785168
2•tosh•37m ago•0 comments

Building Scalable Control Planes

https://www.allthingsdistributed.com/2026/08/on-building-scalable-control-planes.html
1•janvdberg•39m ago•0 comments

Google Search AI leaks ACTUAL name of a game's upcoming character

https://www.reddit.com/r/IndieDev/comments/1vfwuf5/a_player_found_out_the_actual_name_of_my_games/
3•erremerre•40m ago•0 comments
Open in hackernews

Past, Present, and Future of (Ruby) Sorbet Type Syntax

https://blog.jez.io/history-of-sorbet-syntax/
5•ksec•1y ago

Comments

Lio•1y ago
This is a great write up and I think the proposal to make rbs-inline type comments avaible at runtime is an execellent solution.

I can definitely see the value runtime type checking in addition to static analysis. For my own work I've done this either using pattern matching:

  class User
    def initialize(name:, age:)
      String => name
      Integer => age
      @name = name
      @age = age
    end
  end
but the error messages aren't great IMHO. It's pretty easy though knock up a little type checker in a very small module and prepend it. I have support for union types and typed arrays in mine.

I actually wasn't aware of the improvements they've made over the years to Sorbet. It think that's probably what I'll migrate to now.

Lio•1y ago
I'd add, I'm surprised this this post isn't added to the "official" Sorbet blog. It really is a great post and very positive for the future of the project.

https://sorbet.org/blog/