My AI skeptic friends are all nuts (fly.io)

2340 points by tabletcorry 6d ago 2818 comments

Self-hosting your own media considered harmful according to YouTube (jeffgeerling.com)

1610 points by DavideNL 3d ago 750 comments

Bill Atkinson has died (daringfireball.net)

1559 points by romanhn 2d ago 272 comments

The time bomb in the tax code that's fueling mass tech layoffs (qz.com)

1529 points by booleanbetrayal 5d ago 929 comments

OpenAI slams court order to save all ChatGPT logs, including deleted chats (arstechnica.com)

1114 points by ColinWright 4d ago 933 comments

The last six months in LLMs, illustrated by pelicans on bicycles (simonwillison.net)

890 points by swyx 1d ago 223 comments

Cloudlflare builds OAuth with Claude and publishes all the prompts (github.com)

880 points by gregorywegory 7d ago 518 comments

FFmpeg merges WebRTC support (git.ffmpeg.org)

870 points by Sean-Der 5d ago 197 comments

If you are useful, it doesn't mean you are valued (betterthanrandom.substack.com)

845 points by weltview 7d ago 355 comments

Tell HN: Help restore the tax deduction for software dev in the US (Section 174)

824 points by dang 3h ago 338 comments

IRS Direct File on GitHub (chrisgiven.com)

735 points by nickthegreek 5d ago 317 comments

A proposal to restrict sites from accessing a users’ local network (github.com)

673 points by doener 5d ago 377 comments

How to post when no one is reading (jeetmehta.com)

664 points by j4mehta 7d ago 252 comments

Quarkdown: A modern Markdown-based typesetting system (github.com)

662 points by asicsp 6d ago 273 comments

Merlin Bird ID (merlin.allaboutbirds.org)

626 points by twitchard 5d ago 214 comments

Cursor 1.0 (cursor.com)

608 points by ecz 4d ago 466 comments

Why I wrote the BEAM book (happihacking.com)

607 points by lawik 5d ago 151 comments

Deep learning gets the glory, deep fact checking gets ignored (rachel.fast.ai)

597 points by chmaynard 5d ago 157 comments

How we decreased GitLab repo backup times from 48 hours to 41 minutes (about.gitlab.com)

578 points by nayak 3d ago 243 comments

The impossible predicament of the death newts (crookedtimber.org)

568 points by bdr 4d ago 182 comments

Covert web-to-app tracking via localhost on Android (localmess.github.io)

534 points by sebastian_z 6d ago 337 comments

Meta: Shut down your invasive AI Discover feed (mozillafoundation.org)

530 points by speckx 3d ago 224 comments

Show HN: Kan.bn – An open-source alterative to Trello (github.com)

508 points by henryball 7d ago 219 comments

Tesla seeks to guard crash data from public disclosure (reuters.com)

506 points by kklisura 4d ago 430 comments

Show HN: Air Lab – A portable and open air quality measuring device (networkedartifacts.com)

484 points by 256dpi 4d ago 185 comments

Kagi Reaches 50k Users (kagi.com)

483 points by tigroferoce 15h ago 288 comments

EU Commission refuses to disclose authors behind its mass surveillance proposal (old.reddit.com)

482 points by nickslaughter02 6d ago 280 comments

Self-Host and Tech Independence: The Joy of Building Your Own (ssp.sh)

476 points by articsputnik 2d ago 230 comments

My experiment living in a tent in Hong Kong's jungle (corentin.trebaol.com)

471 points by 5mv2 2d ago 210 comments

Washington Post's Privacy Tip: Stop Using Chrome, Delete Meta Apps (and Yandex) (tech.slashdot.org)

462 points by miles 2d ago 308 comments

Falsehoods programmers believe about aviation (flightaware.engineering)

456 points by cratermoon 2d ago 194 comments

Prompt engineering playbook for programmers (addyo.substack.com)

455 points by vinhnx 5d ago 169 comments

Joining Apple Computer (2018) (folklore.org)

439 points by tosh 1d ago 119 comments

Researchers develop ‘transparent paper’ as alternative to plastics (japannews.yomiuri.co.jp)

435 points by anigbrowl 2d ago 277 comments

The Illusion of Thinking: Strengths and limitations of reasoning models [pdf] (ml-site.cdn-apple.com)

435 points by amrrs 3d ago 258 comments

The Right to Repair Is Law in Washington State (eff.org)

433 points by doener 5d ago 169 comments

Convert photos to Atkinson dithering (gazs.github.io)

427 points by nvahalik 1d ago 53 comments

Getting Past Procrastination (spectrum.ieee.org)

422 points by WaitWaitWha 2d ago 169 comments

Google restricts Android sideloading (puri.sm)

419 points by fsflover 4d ago 341 comments

Building supercomputers for autocrats probably isn't good for democracy (helentoner.substack.com)

412 points by rbanffy 22h ago 225 comments

(On | No) Syntactic Support for Error Handling (go.dev)

412 points by henrikhorluck 6d ago 577 comments

Ask HN: How do I learn robotics in 2025?

408 points by srijansriv 7d ago 99 comments

FSE meets the FBI (blog.freespeechextremist.com)

378 points by 1337p337 18h ago 123 comments

A year of funded FreeBSD development (daemonology.net)

376 points by cperciva 3d ago 123 comments

DiffX – Next-Generation Extensible Diff Format (diffx.org)

373 points by todsacerdoti 5d ago 152 comments

Ask HN: Who is hiring? (June 2025)

369 points by whoishiring 7d ago 489 comments

Builder.ai Collapses: $1.5B 'AI' Startup Exposed as 'Indians'? (ibtimes.co.uk)

363 points by healsdata 6d ago 257 comments

Cockatoos have learned to operate drinking fountains in Australia (science.org)

359 points by pseudolus 5d ago 154 comments

The iPhone 15 Pro’s Depth Maps (tech.marksblogg.com)

357 points by marklit 5d ago 93 comments

Autonomous drone defeats human champions in racing first (tudelft.nl)

352 points by picture 4d ago 327 comments

Reasoning LLMs Guide for Devs

3 omarsar 1 6/9/2025, 2:26:09 PM promptingguide.ai ↗

Comments (1)

incomingpain · 5h ago

chatgpt, perpelxity, grok website all rapidly answered correctly. As expected.

claude 3.5 sonnet got it wrong. Came up with 4227.

claude 3.7 sonnet got it right with 5117

claude 4 sonnet also got it wrong. 4227

Gemini 2.5 pro got it wrong with 4227

GPT-4o got it right and wrong answering both 4227 and 5117? It found the right answer on the internet, the code was wrong. Interesting because my free chatgpt in first line is o4-mini and got it right?

deepseek r1 qwen3 8b super failed.

qwen2.5 coder 7b instruct came up with some code which correctly runs and gives me 5117, but the answer was 4291?