news newest ask show jobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Idiomatic Koru Kernels Match Hand-Specialized C

https://www.korulang.org/blog/idiomatic-kernels-match-specialized-c

1•nemoniac•7m ago•0 comments

How to emotionally grasp the risks of AI Safety

https://www.lesswrong.com/posts/jPBmCxpFQzhypeTpg/how-to-emotionally-grasp-the-risks-of-ai-safety

2•joozio•12m ago•0 comments

Honest AI tool reviews. Two good things. One bad thing

https://goodgoodbad.com/

1•colors__•18m ago•0 comments

I implemented Gemini Live Chat as a demo

https://99helpers.com/tools/live-chat

1•nickk81•19m ago•1 comments

Interactive Proofs [video]

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

1•downboots•23m ago•0 comments

Approved Is Not Paid

https://doap.metal.bohyen.space/blog/post/approved-is-not-paid/

1•nanacnote•25m ago•0 comments

Show HN: Real-time 3D London Underground trains using the TFL API

https://www.minilondon3d.xyz/

1•kiranbaby14•29m ago•0 comments

Show HN: Buddy Board – Pokédex-style trading cards for Claude Code companions

2•Tanayk07•35m ago•0 comments

SereneCode – Formal verification framework for AI-generated Python code

https://github.com/helgster77/serenecode

1•doktorice•36m ago•0 comments

Show HN: Tusk for macOS and Gnome

https://shapemachine.xyz/tusk/

1•factorialboy•36m ago•0 comments

GEOscore – Check if AI assistants can find and cite your website

https://geoscore-tawny.vercel.app

1•albanthaumur•38m ago•0 comments

I Made Tinder but for Australian News

https://oznews.live

1•arshmansuri•39m ago•1 comments

Real-time global illumination with WebGPU

https://juretriglav.si/surfel-based-global-illumination-on-the-web/

1•juretriglav•49m ago•0 comments

Apple: Embarrassingly Simple Self-Distillation Improves Code Generation

https://arxiv.org/abs/2604.01193

4•Anon84•50m ago•0 comments

Garry's List – the Y Combinator CEO's civic engagement project

https://garryslist.org/

1•sph•57m ago•0 comments

In California, wildfires are prevented by crews of unlikely firefighters: goats

https://www.npr.org/2023/08/10/1192905277/goat-grazing-california-wildfire-prevention

2•thunderbong•58m ago•0 comments

DriftNote – Podcast notes that talk back

https://www.driftnote.net/

1•LifeOfKP•1h ago•1 comments

Tests confirm super-fast charging for first solid-state-battery e-moto

https://newatlas.com/motorcycles/donut-lab-solid-state-battery-charge-test/

1•breve•1h ago•0 comments

The Clock

https://blog.senko.net/the-clock

2•senko•1h ago•1 comments

CDC and health groups spent millions in ads on sites flagged for misinformation

https://www.cidrap.umn.edu/anti-science/cdc-health-groups-spent-millions-buy-ads-websites-flagged...

1•giuliomagnifico•1h ago•0 comments

Angine de Poitrine – Full Performance [video]

https://www.youtube.com/watch?v=0Ssi-9wS1so

1•vitto_gioda•1h ago•0 comments

PDS Mac OLM File to PST Converter

https://www.perfectdatasolutions.com/en/olm/olm-to-pst-converter.html

1•tieanderson•1h ago•0 comments

A brief history of instant coffee

https://worksinprogress.co/issue/a-brief-history-of-instant-coffee/

3•admp•1h ago•0 comments

GPUs vs. TPUs: Decoding the Powerhouses of AI

https://www.savvycanary.com/gpus-vs-tpus-decoding-the-powerhouses-of-ai/

1•car•1h ago•0 comments

Being the Human in the Loop – Kevlin Henney

https://www.youtube.com/watch?v=vpYJMr1pJRY

1•mcp_•1h ago•0 comments

Emotion Concepts and Their Function in a Large Language Model

https://transformer-circuits.pub/2026/emotions/index.html

3•Anon84•1h ago•0 comments

Trinity-Large-Thinking: Open-source 398B MoE (13B active) for agentic tasks

https://firethering.com/trinity-large-thinking-open-source-agent-model/

3•steveharing1•1h ago•0 comments

Trying to Be Responsible

https://chatgpt.com/s/t_69d0ca9364ac8191868c2850d26305aa

1•aljgz•1h ago•1 comments

Every dependency you add is a supply chain attack waiting to happen

https://benhoyt.com/writings/dependencies/

1•ingve•1h ago•0 comments

Engineering a Better Java Build Tool [video]

https://www.youtube.com/watch?v=OtsJ902k458

1•lihaoyi•1h ago•0 comments

Open in hackernews

I solved almost all of free problems on LeetCode using AI

https://old.reddit.com/r/leetcode/comments/1krg7by/i_solved_almost_all_of_free_problems_on_leetcode/

3•tevlon•10mo ago

Comments

cratermoon•10mo ago

I'm sure all the free problems from LeetCode are in the training sets, so I don't think this is much of a flex.

tevlon•10mo ago

i was thinking the same. And i am sure it is in the training set, but i couldn't solve 10 hard problems. Also, most of the hard problems, the LLMs weren't able to solve in one shot. How do you explain this?

cratermoon•10mo ago

There are many more correct answers to the easy problems than to the hard ones. Thus the LLM is more likely to hit upon the correct completion for them.

tevlon•10mo ago

makes sense

whobre•10mo ago

Alright, but how can this exercise help them pass an interview? The point of leetcode is to memorize the solutions so one can do well in an interview; yes, I agree it’s ridiculous.