Hacker News: kang

New comment by kang in "OpenAI is well positioned to fast-follow Jev"

kang — Tue, 22 Sep 2026 17:17:35 +0000

except the output is indistinguishable from hallucination.

New comment by kang in "India has paved the way for charging merchants a fee on UPI transactions"

kang — Tue, 18 Aug 2026 19:13:13 +0000

both upi & cards are backwards. canada like etransfer to phone to email is the best modern solution so far.

New comment by kang in "What is a product?"

kang — Fri, 07 Aug 2026 08:07:37 +0000

A product is something consumable.

New comment by kang in "São Paulo resident transforms degraded area into urban forest"

kang — Fri, 07 Aug 2026 06:05:40 +0000

ive had hard time finding place to plant trees, can someone share if they of any 'forgotten lands'

New comment by kang in "qm – Multiplayer agent harness for work"

kang — Fri, 31 Jul 2026 21:33:07 +0000

this one is trying to cashin on the 'multiplayer' keyword after the buzz that 'no one has cracked multiplayer yet'. well the ui sucks and is not it.

New comment by kang in "Ask HN: What are tools you have made for yourself since the advent of AI?"

kang — Mon, 08 Jun 2026 22:27:05 +0000

Android browser wrapper that can download any video, audio, text

New comment by kang in "Bricks and Minifigs Stole a Man's $200k Lego Collection"

kang — Fri, 29 May 2026 12:15:40 +0000

No one talking about the police brutality & governance issues.

New comment by kang in "Building Pi with Pi"

kang — Mon, 25 May 2026 11:21:04 +0000

clanking is just a sound made by robot's metal. not derogatory and isn't even meant to be used for agents but robots.

New comment by kang in "Constraint Decay: The Fragility of LLM Agents in Back End Code Generation"

kang — Mon, 25 May 2026 11:13:04 +0000

hehe, this is by design. next model needs to eat your natural language

New comment by kang in "We see something that works, and then we understand it"

kang — Sun, 10 May 2026 07:34:26 +0000

this misunderstands whats thinking is ..

> Thinkism sets aside practice and experience

thinking succeeds experience & precedes practise, its not apart from it

New comment by kang in "We see something that works, and then we understand it"

kang — Sun, 10 May 2026 07:30:09 +0000

try replacing the word with 'thinking'

New comment by kang in "A recent experience with ChatGPT 5.5 Pro"

kang — Sun, 10 May 2026 07:28:42 +0000

Verification & solution generation are both part of problem generation & defining the passing test - judgement.

New comment by kang in "Using Claude Code: The unreasonable effectiveness of HTML"

kang — Sun, 10 May 2026 04:03:25 +0000

orgdown is better than markdown which is better than markup, except the text ain't hyper. i've been working on this since feb & have reached xanadu

New comment by kang in "A recent experience with ChatGPT 5.5 Pro"

kang — Sat, 09 May 2026 23:04:56 +0000

Verification on its own is not research, but judgement is research.

"Hey, Prove something a machine can't", sure I can't, "Hey, Say something worth proving & judge it well", ah, now I might have a few unique observation/ideas/curiosities/problems from my having being a human.

Imo, the feeling of intelligence or the process of originality(originativity) test for ai is subjective & is coming down to 4 paths: novel relative to a reference class, valuable within a domain, counterfactually sensitive to internal state and environment, and revisable through learning.

New comment by kang in "Light without electricity? Glowing algae could make it possible"

kang — Sat, 09 May 2026 10:41:12 +0000

Unlike artificial carbon capture, natural carbon capture like algae here become insect/worm/bird feed or manure/coal.

New comment by kang in "A recent experience with ChatGPT 5.5 Pro"

kang — Sat, 09 May 2026 10:34:00 +0000

> The lower bound for contributing to mathematics will now be to prove something that LLMs can’t prove, rather than simply to prove something that nobody has proved up to now and that at least somebody finds interesting.

5.5pro is amazing but this implication might not be true & is the core argument of this piece.

AI will prove all sort of things - interesting, boring & incorrect.

To sort it will be the task of the PhD.

New comment by kang in "Cursor Camp"

kang — Thu, 30 Apr 2026 03:52:00 +0000

I saw this experiment decades ago on the internet and it was to a music concert, i always wanted to do a cursor moshpit

New comment by kang in "A playable DOOM MCP app"

kang — Tue, 28 Apr 2026 20:23:50 +0000

somebody did this a month ago https://www.youtube.com/watch?v=fdbXNWkpPMY

i am increasingly going schizo, where every single thing I post/see posted gets copied and karma farmed on social media. further, any novelty I share with an llm gets eaten/absorbed by the harness as a feature.

New comment by kang in "An update on recent Claude Code quality reports"

kang — Sun, 26 Apr 2026 03:32:03 +0000

You are right, I was wrong in my understanding there. It stemmed from my own implementation; an inference often wrote extra data such as tool call, so I was using it to preserve relevant information alongwith desired output, to be able to throw away the prompt every time. I realize inference caching is one better way (with its pros and cons).

New comment by kang in "GPT‑5.5 Bio Bug Bounty"

kang — Sat, 25 Apr 2026 21:01:38 +0000

this economic model works for all 'bounty' related work