frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

NTSB Animation of Flight 5342 2025 Potomac River mid-air collision [video]

https://www.youtube.com/watch?v=LJ10ZOcWuC4
1•antongribok•58s ago•0 comments

Why is the app slow on Pixel 7?

https://binarysky.se/blog/2026/01/why-is-the-app-slow-on-pixel-7/
1•lindskogen•1m ago•0 comments

Sum, Product: A simple-to-understand mathematical paradox on meta-cognition

https://magarshak.com/blog/sum-product-and-the-limits-of-meta-knowledge/
1•EGreg•2m ago•0 comments

40 years later, a new look at lessons from the Challenger disaster

https://www.washingtonpost.com/politics/2026/01/27/challenger-space-shuttle-disaster-40-years/
1•bookofjoe•3m ago•1 comments

Chasing 6 TB/s: an MXFP8 quantizer on Blackwell

https://blog.fal.ai/chasing-6-tb-s-an-mxfp8-quantizer-on-blackwell/
1•amrrs•3m ago•0 comments

How-Dirty-Marketing-Works

https://how-dirty-marketing-works.onrender.com/
1•yashsm01•4m ago•0 comments

We Built Our IVR Scraper (and What Broke Along the Way)

https://phonesupported.dev/blog/how-we-built-our-ivr-scraper-and-what-broke-along-the-way/
1•fast_kalyan•5m ago•0 comments

Chrome, Edge Extensions Caught Stealing ChatGPT Sessions

https://www.securityweek.com/chrome-edge-extensions-caught-stealing-chatgpt-sessions/
1•Bender•7m ago•0 comments

Six JavaScript zero-day bugs lead to fears of supply chain attack

https://www.scworld.com/news/six-javascript-zero-day-bugs-lead-to-fears-of-supply-chain-attack
1•Bender•8m ago•0 comments

The sad and self-inflicted decline of the Washington Post, in one chart

https://www.natesilver.net/p/the-sad-and-self-inflicted-decline
2•JumpCrisscross•8m ago•0 comments

Coding with AI without giving up power

https://medium.com/@sig.segv/coding-with-ai-without-giving-up-power-e0c6ca257ad9
1•fwef64•9m ago•0 comments

Supreme Court to decide how 1988 videotape privacy law applies to online video

https://arstechnica.com/tech-policy/2026/01/supreme-court-to-decide-how-1988-videotape-privacy-la...
2•Bender•9m ago•0 comments

Units of measure in the KCL CAD language

https://www.ncameron.org/blog/kcl-part-1-units/
1•fanf2•9m ago•0 comments

Fun 0.37.62 – The programming language that makes you have fun

https://fun-lang.xyz/2026/01/25/announcing-fun-0.37.62/
1•hanez•10m ago•1 comments

GameStop's original bull is back. CEO Ryan Cohen has Berkshire-like ambitions

https://www.msn.com/en-us/money/savingandinvesting/gamestop-s-original-bull-is-back-ceo-ryan-cohe...
1•petethomas•11m ago•0 comments

How many users can a $50K AI workstation serve? Benchmark data

https://old.reddit.com/r/LocalLLaMA/comments/1qorbdk/dual_rtx_pro_6000_workstation_with_115tb_ram/
2•Blue_Cosma•12m ago•0 comments

AI gives cleaning instructions [video]

https://www.youtube.com/shorts/TP_p5YnRH00
1•6510•12m ago•0 comments

"IG is a drug": Internal messages may doom Meta at social media addiction trial

https://arstechnica.com/tech-policy/2026/01/tiktok-settles-hours-before-landmark-social-media-add...
4•mikestew•14m ago•0 comments

Kimi K2.5 1T runs on 2 M3 Ultras with mlx-lm in it's native precision

https://twitter.com/awnihannun/status/2016221496084205965
2•jeudesprits•16m ago•0 comments

Asteroid 2024 YR4 Has a 4% Chance of Hitting the Moon

https://www.universetoday.com/articles/asteroid-2024-yr4-has-a-4-chance-of-hitting-the-moon-heres...
1•belter•16m ago•0 comments

Hacker News Slop

https://www.hnslop.com/
3•jshchnz•17m ago•0 comments

Updated LLM Benchmark (Gemini 3 Flash)

https://entropicthoughts.com/updated-llm-benchmark
2•surprisetalk•17m ago•1 comments

Quality, Quantity, and Effective Iteration

https://www.autodidacts.io/iteration/
1•surprisetalk•18m ago•0 comments

George Washington's Teeth

https://en.wikipedia.org/wiki/George_Washington%27s_teeth
1•surprisetalk•18m ago•0 comments

Opentrees.org (2024)

https://opentrees.org/#pos=1/-37.8/145
2•surprisetalk•18m ago•0 comments

A B-52 Crash in Greenland Became One of the Cold War's Worst Nuclear Accidents

https://www.military.com/daily-news/investigations-and-features/2026/01/16/how-b-52-crash-greenla...
2•belter•19m ago•0 comments

It is incorrect to "normalize" // in HTTP URL paths

https://runxiyu.org/comp/doubleslash/
1•todsacerdoti•19m ago•0 comments

Show HN: Duperr – On-device voice cloning using Qwen3-TTS for iOS

https://testflight.apple.com/join/98d1hB6U
1•ahampt•20m ago•0 comments

Pornhub to restrict access for UK users

https://www.bbc.co.uk/news/articles/czr428rxg57o
2•midlander•21m ago•0 comments

What would GitHub Actions look like if you designed it today?

https://www.rwx.com/blog/what-would-github-actions-look-like-if-you-designed-it-today
2•dan_manges•23m ago•0 comments
Open in hackernews

Show HN: Agent Composer – Create your own AI rocket scientist agent

https://demo.contextual.ai/
10•jayc481•1h ago
Hey HN! We launched a thing today, and built a cool demo that I'm excited to share with the community.

This tool creates AI agents easily and can handle some really technically complex work. I whipped up this rocket scientist agent in our tool in 10 minutes. I asked a couple of aerospace engineer friends what they thought, and they found it useful for their real work, which is why I want to share it here. Here's the demo https://demo.contextual.ai/ and an overview https://docs.contextual.ai/examples/rocket_science

Would love to see technical HN folks and their experience with our tool (and whether it would be useful in your work). Happy to go deep on architecture, retrieval strategies, and lessons learned.

Comments

binarysolo•1h ago
Is there something a little more general/accessible that you guys have as a demo? It looks super cool but I'm not a rocket scientist so I can't tell how good it is... (am def into agentic AI and am mildly involved in this in the ecom space).
ninalopatina•1h ago
We have our agent on our docs page! (Hi I also worked on this with Jay) https://docs.contextual.ai/ - if you type a question into "ask AI", it's powered by agent composer just like the rocket demo. And we actually have a blog about an ecom agent coming out soon, it was one of our test cases due to a large, public dataset available in this space.
ninalopatina•1h ago
The inline citations are really convenient