Hacker News: drdrey

New comment by drdrey in "Bend – a language that blocks AI mistakes via proof and runs on GPUs"

drdrey — Fri, 18 Sep 2026 00:26:27 +0000

writing specs is hard. For instance the easiest way to comply with "player should never reach the flag" is to disable movement completely, so then you have to specify "player should never reach the flag while still being allowed to move" and so on (liveness). It's not practical for most programs

New comment by drdrey in "iPhone 18 Pro and iPhone 18 Pro Max"

drdrey — Thu, 10 Sep 2026 01:30:38 +0000

you could do edits like crops in a way that preserves the signature using ZK proofs

New comment by drdrey in "Oceans hit highest temperature on record"

drdrey — Mon, 24 Aug 2026 21:31:43 +0000

1.5% of electricity, not energy

that would be 0.3% of global energy use

New comment by drdrey in "Mea Culpa – Dark Hours"

drdrey — Sun, 09 Aug 2026 16:35:10 +0000

"that's on me" just sounds like Claude is doing the talking

New comment by drdrey in "Ten advances in mathematics and theoretical computer science"

drdrey — Sat, 01 Aug 2026 15:37:53 +0000

> The results were achieved by an internal version of Astra, our next major model.

New comment by drdrey in "The Maxwell Conjecture Is False (GPT 5.6 Sol)"

drdrey — Fri, 31 Jul 2026 16:08:37 +0000

of course it does, but it's still the best thing we have

New comment by drdrey in "Hacker wipes Romania's land registry database"

drdrey — Mon, 20 Jul 2026 14:27:42 +0000

fires? floods? documents getting lost/misplaced/stolen/destroyed?

New comment by drdrey in "GPT-5.6"

drdrey — Fri, 10 Jul 2026 02:13:13 +0000

the problems are general and abstract, domain specific knowledge and memorization don't help. Figuring out the rules, the goal, the controls, and how to solve in a reasonable budget all indicate some level of general ability.

New comment by drdrey in "Please Do Not Vibe Fuck Up This Software"

drdrey — Sun, 31 May 2026 04:14:49 +0000

Counterpoint: don't read the comments

New comment by drdrey in "The Last Technical Interview"

drdrey — Sat, 30 May 2026 02:03:43 +0000

> And it can pay for itself twice: once in real work shipped, and once again in something else you could probably use more of, which is gravity.

what?

New comment by drdrey in "Incident with Pull Requests, Issues, Git Operations and API Requests"

drdrey — Wed, 27 May 2026 13:45:35 +0000

as someone who doesn't know how to get better design out of LLMs, can you elaborate?

New comment by drdrey in "Defeating Git Rigour Fatigue with Jujutsu"

drdrey — Sun, 24 May 2026 23:43:23 +0000

I have finally embraced squashing PRs and realized I wasted my youth trying to write Good Commits.

New comment by drdrey in "Shunning AI is the human choice"

drdrey — Thu, 21 May 2026 17:06:53 +0000

it's tribal, it feels good to rally against a common enemy

New comment by drdrey in "Cybersecurity looks like proof of work now"

drdrey — Wed, 15 Apr 2026 21:55:16 +0000

good luck formally verifying everything

New comment by drdrey in "Tech employment now significantly worse than the 2008 or 2020 recessions"

drdrey — Sat, 07 Mar 2026 14:37:11 +0000

that ship has sailed

New comment by drdrey in "I’m joining OpenAI"

drdrey — Mon, 16 Feb 2026 15:05:52 +0000

for me it's that it works remotely and my kids can access it over Discord

New comment by drdrey in "Composing APIs and CLIs in the LLM era"

drdrey — Fri, 23 Jan 2026 00:59:16 +0000

this seems like what MCPs should have been from the beginning. If you think MCPs are better, can you explain why?

New comment by drdrey in "GitHub should charge everyone $1 more per month to fund open source"

drdrey — Wed, 14 Jan 2026 17:04:31 +0000

the problem with any approach like this based on usage metrics is that it will be abused to death

New comment by drdrey in "Claude Code CLI was broken"

drdrey — Wed, 07 Jan 2026 21:16:08 +0000

...but is it good?

New comment by drdrey in "GPT-5.2-Codex"

drdrey — Fri, 19 Dec 2025 16:01:33 +0000

have you tried adjusting the reasoning effort?