fp.

Code is abundant. Judgment is scarce

https://twitter.com/harjotsgill/status/2087532414025216348

1•TheAnkurTyagi•1m ago•0 comments

The 108-Degree Eviction

https://www.theatlantic.com/health/2026/08/vegas-heat-wave-homelessness/688160/

1•rafram•1m ago•0 comments

Fail Faster

https://derickrethans.nl/fail-faster.html

1•speckx•1m ago•0 comments

Performance Impact of Alignment

https://eme64.github.io/blog/2026/01/12/Alignment-Performance.html

1•signa11•2m ago•0 comments

Show HN: Dynobox – A test runner for AI agent skills and workflows

https://github.com/dynobox/dynobox

1•bhkdotdev•2m ago•0 comments

Pixel Watch 5

https://blog.google/products-and-platforms/devices/pixel/pixel-watch-5/

1•ortusdux•2m ago•0 comments

Google Pixel Tag

https://blog.google/products-and-platforms/devices/pixel/google-pixel-tag/

1•ortusdux•3m ago•0 comments

C++26: Std:Indirect

https://www.sandordargo.com/blog/2026/08/12/cpp26-indirect

2•signa11•3m ago•0 comments

Decrypting my Flume water monitor's MQTT traffic with a passive relay

https://lithostech.com/2026/08/decrypting-flume-water-monitor-traffic/

1•stevecrozz•5m ago•0 comments

CodeRabbit raises a $143M Series C at a $1.5B valuation

https://www.coderabbit.ai/blog/introducing-agentic-change-management

2•juanpflores•5m ago•0 comments

We tested our new Voice Isolation across 10 STT engines. Word errors dropped 46%

https://krisp.ai/blog/voice-isolation-2-5/

1•davitb•6m ago•0 comments

Different SAST/DAST scanners only share %1.8 of common findings on same source

https://blog.audn.ai/posts/audn-vs-codex-vs-aikido-juice-shop-comparison

1•boratac•6m ago•0 comments

DeepSeek-V4-Pro-0813

1•uneven9434•6m ago•0 comments

Music notes are no longer the bottleneck (ca. 1600)

https://bsky.app/profile/jeff.auriemma.xyz/post/3msvhsitaf22m

2•jdauriemma•6m ago•0 comments

DosTips, Windows scripting knowledge trove, scraped to death

https://www.dostips.com/

1•iqml2568•6m ago•0 comments

RL Environments Explained: How AI Agents Learn Real-World Work [video]

https://www.youtube.com/watch?v=a00xIn5kwhM

1•ronfriedhaber•7m ago•0 comments

There is no escaping the solar system, not at speed anyway

https://disassociated.com/no-escaping-solar-system-not-speed/

1•Brajeshwar•8m ago•0 comments

Grok 4.6 (High) Intelligence, Performance and Price Analysis

https://artificialanalysis.ai/models/grok-4-6

2•theanonymousone•11m ago•0 comments

Footage Shows Dolphins Hunting with Shells Off the East Coast of Australia

https://www.sciencealert.com/dolphins-filmed-hunting-with-shells-and-possibly-passing-the-skill-t...

1•thunderbong•11m ago•0 comments

Where Has All the Testosterone Gone?

https://www.theatlantic.com/family/2026/08/testosterone-decline-theories/688247/

2•fortran77•11m ago•0 comments

Show HN: My pocket-sized Personal IoT device. Data-local by default

3•AJTSheppard•12m ago•0 comments

DeepSeek V4 Pro 0813

https://openrouter.ai/deepseek/deepseek-v4-pro-0813

6•explosion-s•12m ago•1 comments

IO Factory: Simulating AI-Enabled Influence Campaigns at Scale [pdf]

https://arxiv.org/abs/2608.10920

1•andrybak•12m ago•0 comments

YouTuber Hank Green says his AI usage is 'not healthy'

https://techcrunch.com/2026/08/01/youtuber-hank-green-says-his-ai-usage-is-not-healthy/

2•abnercoimbre•13m ago•2 comments

Show HN: Aphorio – 3,250,000 WebSocket sends in 0.002 seconds (Ryzen 9 9900X)

https://github.com/prettydiff/aphorio

1•austin-cheney•13m ago•0 comments

Math identity that applied at model-load time, makes MLA decode at 71,000 tok/s

https://efficientagent.substack.com/p/the-mla-decode-speedup-hiding-in

1•bnayak25•14m ago•0 comments

Putting Claude's Watermarking to the Test

https://dreasays.substack.com/p/putting-claudes-watermarking-to-the

1•gojkoa•14m ago•0 comments

Archer OS: draft spec for AI agents operating apps under OS authority

https://github.com/coachpato/archer-os

2•Billie_Archer•17m ago•0 comments

Remind HN: Perseid meteor shower peaks tonight

1•cjbarber•17m ago•0 comments

NIST RFI for modernizing the NVD in the wake of AI

https://www.federalregister.gov/documents/2026/08/12/2026-16371/request-for-information-rfi-on-mo...

2•MattSayar•18m ago•0 comments

Open in hackernews

SHOW HN: Why.com – a curiosity engine back from 1999

https://why.com

1•rasheed106•1h ago

Comments

rasheed106•1h ago

WHY started in 1999 with a small team from Harvard. Raised $4m, got hit during dot com bust and went dark for over 25 years.

The homepage is still here: https://web.archive.org/web/20001019031120/http://why.com/in...

The original founder is back and I built why.com around a simple loop: one answer, three new questions.

The foundation model is at epicenter and orchestration was my taste so tear it apart.

I’m curious.