frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Ask HN: What should I know before hosting eprints?

1•logicallee•1m ago•0 comments

Qwen3.8 27B Runs 200 tok/s on a Single RTX 5090

https://twitter.com/sgl_project/status/2088281320422322413
1•unseenmars•3m ago•0 comments

An easy way to build self-diagnosing and self-learning AI agents in production

https://www.youtube.com/watch?v=2x0BKQBMbF4
1•supportm•4m ago•1 comments

Castform now supports supervised finetuning

https://castform.com/blog/sft-launch/
1•kumama•5m ago•0 comments

UI is hell: four-function calculators

https://lcamtuf.substack.com/p/ui-is-hell-four-function-calculators
1•jandeboevrie•7m ago•0 comments

Capyreader: A smallish Android RSS reader

https://github.com/jocmp/capyreader
1•thunderbong•8m ago•0 comments

Independent Internet Forums

https://indieforums.linklists.net/
1•rdmuser•10m ago•0 comments

Comparing React and Htmx for Building Chatbots

https://carlkolon.com/2026/08/14/htmx-react-chat/
1•cckolon•12m ago•0 comments

The Making of Pro Office Calculator, Part 1 – Fragment-Based Architecture

https://deadbeef.io/procalc_making_of_1
1•rjinman•12m ago•0 comments

When facts change, should AI memory change?

https://github.com/Anchorstate-Lab/GMR
1•Zongming•13m ago•0 comments

Turning molecules into reliable electronic devices

https://news.mit.edu/2026/turning-molecules-into-reliable-electronic-devices-0803
1•gmays•14m ago•0 comments

Human brain cells do next-token prediction

https://parasma.com/news/human-brain-cells-do-next-token-prediction
1•alach11•16m ago•0 comments

Apple trains its own AI model for China market with Alibaba's support, sources

https://www.reuters.com/business/retail-consumer/apple-trains-its-own-ai-model-china-market-with-...
3•newusertoday•16m ago•0 comments

Trust Boundary Collapse: Two Campaigns That Bypassed Artifact Inspection

https://lateos.ai/articles/trust-boundary-collapse-teampcp-clickfix
1•leochong•17m ago•0 comments

Multi-hop reasoning collapses by 1M tokens; with a learning system it doesn't

https://zenodo.org/records/21937795
1•lvrzhn•17m ago•0 comments

Rustc_codegen_JVM: Compile Rust to Java bytecode and runnable JARs

https://github.com/IntegralPilot/rustc_codegen_jvm
1•fanf2•17m ago•0 comments

AI Thinks the Internet Is Reddit. We Have 8,616 Answers Proving It

https://growtika.com/blog/reddit-ai-visibility-research
3•Growtika•18m ago•2 comments

1,1 mld USD dla River AI: Co oznacza rewolucja spersonalizowanego AI?

https://neurise.io/blog-auto-1-1-mld-usd-dla-river-ai-rewolucja-spersonalizowanego-ai.html
1•mbarczynski•18m ago•0 comments

Can Boeing Be Forced to Let a Blind Machinist Build Planes? New Lawsuit Says Yes

https://viewfromthewing.com/can-boeing-be-forced-to-let-a-blind-machinist-build-planes-a-new-laws...
1•crescit_eundo•19m ago•1 comments

LLM Web Search Benchmarks: More Searches Beat a Better Search Engine

https://openrouter.ai/blog/announcements/web-search-benchmark/
1•DGAP•20m ago•0 comments

Earn yield through machine learning vault on DeFi

https://twitter.com/macbrennan_cc/status/2088295071024034226
1•panterafii•22m ago•1 comments

150 years of Gibbs' phase rule

https://www.chemistryworld.com/features/150-years-of-gibbs-phase-rule/4023900.article
2•akshatjiwan•24m ago•0 comments

Predicting short interest from FINRA daily short volume shows 0.414 correlation

https://equibles.com/research/does-daily-short-volume-predict-short-interest
1•daniel3303•24m ago•0 comments

An advanced attacker is targeting Salesforce and ServiceNow instances worldwide

https://www.reco.ai/blog/city-forum-campaign-salesforce-servicenow
1•daretograsp•25m ago•0 comments

PostgreSQL: The database that should be dead but runs the Internet [video]

https://www.youtube.com/watch?v=_CB_Aa2ODeM
1•teleforce•25m ago•1 comments

US Court rules AFP health programs can't import Gilead medications from overseas

https://www.cnbc.com/2026/08/14/court-ruling-gilead-medications-overseas-afps.html
1•ck2•27m ago•1 comments

Agentic AI drops digits, This library catches them before it runs in fintech

https://github.com/insightitsGit/PrismManifest
1•parvaamin•27m ago•0 comments

Desktop Calendar: A Design Journey – The Thunderbird Blog

https://blog.thunderbird.net/2026/08/desktop-calendar-a-design-journey/
1•speckx•29m ago•0 comments

Visual Studio Code 1.133

https://code.visualstudio.com/updates/v1_133
1•tosh•30m ago•0 comments

Nvidia's $500B Plan Envelops Wall Sreett in Its AI Frenzy

https://www.bloomberg.com/news/articles/2026-08-14/nvidia-s-500-billion-plan-envelops-wall-street...
1•petethomas•31m ago•0 comments
Open in hackernews

The Bots Always Find Something

https://faingezicht.com/articles/2026/08/14/the-bots-always-find-something/
2•avyfain•50m ago

Comments

SummSolutions•26m ago
Very similar situation here. Months of agentic coding and seem to find problems in almost every test. And every model seems to find different failures. It’s a major issue if you don’t know what’s worth fixing and what isn’t.