frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Apple Just Added Face Unlock to iPhone Duo [video]

https://www.youtube.com/watch?v=P_hmw2EtZxM
1•mgh2•1m ago•0 comments

In the Age of A.I. Slop, a Montebello Metalhead Still Proudly Draws Art by Hand

https://lataco.com/black-metal-punk-album-art
1•cdrnsf•2m ago•0 comments

DOJ to review DoorDash-Instacart merger

https://www.reuters.com/world/doordash-instacart-merger-be-reviewed-by-doj-sources-say-2026-11-16/
1•JumpCrisscross•2m ago•0 comments

Apple Watch and Garmin are so wrong! How many calories do you burn in a day?

https://macrocodex.app/knowledge/rethink/how-many-calories-do-you-burn-in-a-day-apple-watch-and-g...
1•faangguyindia•6m ago•0 comments

Show HN: Win98 PE Interpreter in WebAssembly – StarCraft and Diablo on Phones

https://wine-assembly.berrry.app/
1•vgrichina•7m ago•0 comments

Meta would 'consider' pulling services from Canada if lawful access bill passes

https://www.cbc.ca/lite/story/9.7424473
2•colinprince•9m ago•1 comments

Show HN: Omnistew – 3D multiplayer RTS written in Rust, playable in a browser

https://omnistew.io/
1•omnistew•10m ago•0 comments

The AI Ascent

https://medium.com/@chipmunkworks/the-helicopter-problem-9bfec2cfdfab
1•treelover•12m ago•1 comments

Geographic Data Science with Python

https://geographicdata.science/
1•teleforce•13m ago•0 comments

Claude Is Always Right

2•KenPainter•13m ago•1 comments

Habits PhDs Must Unlearn to Build a Startup [video]

https://www.youtube.com/watch?v=u-8KpXcPYCw
1•gmays•18m ago•0 comments

Claude Mythos slashed software work to onboard AMD GPUs to around three days

https://www.tomshardware.com/tech-industry/artificial-intelligence/anthropic-says-claude-mythos-s...
2•nsoonhui•22m ago•0 comments

Show HN: 1.499x speedup and 28.5% dynamic energy drop on A40 via BF16 tiling

https://cwtractors.com
3•ktractor123•26m ago•0 comments

Are forensic experts biased? Scientist's claims spark outrage – Science – AAAS

https://www.science.org/content/article/forensic-experts-biased-scientists-claims-spark-outrage
1•wglb•33m ago•0 comments

The Weather at Home

https://alexhowells.com/writing/the-weather-at-home/
3•nixgeek•33m ago•0 comments

Show HN: Fundlark-open source tool help early-career researcher find funding

https://github.com/dryang21/Fundlark
1•dryang21•36m ago•0 comments

It's ok that computers are (sometimes) better than us

https://stackoverflow.blog/2026/10/30/it-s-ok-that-computers-are-sometimes-better-than-us/
2•mooreds•37m ago•0 comments

Nvidia rules out Jensen Huang comeback as campaign for his return grows

https://www.reuters.com/legal/transactional/silicon-valley-leaders-back-call-jensen-huangs-return...
2•A_D_E_P_T•38m ago•0 comments

Running coding agents without handing them my credentials

https://blog.antonio.cx/running-agents-with-lend
10•antoniomika•39m ago•2 comments

Friselle: The ingenious Italian bread that can last for a year

https://www.msn.com/en-ca/food-and-drink/recipes/friselle-the-ingenious-italian-bread-that-can-la...
5•BiraIgnacio•39m ago•0 comments

Show HN: Free weekly backups for Supabase Free projects (pg_dump to R2)

https://github.com/liulewis/supabase-free-backup-lite
1•agricalc•44m ago•0 comments

Own your AI, at full speed

https://www.ycombinator.com/launches/XcZ-rightnow-own-your-ai-at-full-speed
5•Jr23_xd•44m ago•0 comments

My Annoying Coworker Claude

https://sharamp.com/personal/claude/
1•BrokenMirror•46m ago•0 comments

Old SaaS Model Is Dead. Software Is a Plug-In. The Hard Part Is Getting Picked

https://www.thestateofai.com/news/saas-is-dead-software-plug-in-ai-assistants
2•Gaishan•46m ago•0 comments

Collisions in Hash Functions

https://collisions.zksecurity.xyz/
1•baby•47m ago•0 comments

The Cyanotype Network

https://network.studiowetware.com/index.html
2•Beachkilla•48m ago•0 comments

TimesFM-3: A zero-shot foundation model for multivariate forecasting

https://research.google/blog/timesfm-3-a-zero-shot-foundation-model-for-multivariate-forecasting/
4•gmays•49m ago•3 comments

Gemini Decision 1 can't separate AI voices made after 2024

https://echo.ls/posts/gemini-decision-ai-voices/
1•maplet•50m ago•0 comments

Ask HN: Can using a VPN get your account shadowbanned or flagged?

1•illidan1000•51m ago•1 comments

Show HN: I made an audio-to-MIDI model 60x more efficient

https://kobi.music/research/dettato
2•hidude562•55m ago•0 comments
Open in hackernews

Claude Is Always Right

2•KenPainter•13m ago
I think I can say, at least for me, why Claude's writing is so freaking awful, even when it's completely correct.

Claude writes technical docs that are true in the same sense that it is true my best friend is a collection of specialized tissues known as "organs" and blah blah blah - and nothing else. Don't ever forget the trailing negation, it is omitted never.

Another take: Claude writes like your best friend describing a great first date. "Ok we were going to meet at 8 - did I mention it was Friday? Right, 8 oclock at the Italian place and no place else. During dinner he accidentally spit out some food, but I thought i was funny! That's because we were laughing so hard after we discovered we were both from Omaha. That was before dinner, during drinks, when we couldn't get rid of Alfred. His place really needs a deep cleaning!"

Perhaps with all of the hope for an LLM-authored movie, we try for something non-linear?

Comments

abraxas•7m ago
It's likely because LLMs by their nature can't backtrack. It will probably improve the output if you tell it to rewrite the response in a linear flow without digressions.