Hacker News: ludamad

New comment by ludamad in "Setting up your spare Mac for Claude Code to control, a step-by-step guide"

ludamad — Sun, 19 Jul 2026 01:22:58 +0000

I don't know what sort of shell scripts you've had to review in your day... I used to get far worse shell scripts, assuredly

New comment by ludamad in "GPT-5.6 Sol Ultra will be in Codex"

ludamad — Mon, 06 Jul 2026 02:01:51 +0000

I imagine this is something like Anthropic's dynamic workflows where a JS file is created to make a little AI harness on the spot

New comment by ludamad in "The Reports of Jim Carrey's Death Are a Failure Mode"

ludamad — Sat, 04 Jul 2026 17:44:19 +0000

Yeah exactly. It's a big circle of references without any hard proof. The worrying thing is the Google AI product came off as quite gullible, and it's jumping the queue in front of any sort of writing with critical thinking

New comment by ludamad in "The Reports of Jim Carrey's Death Are a Failure Mode"

ludamad — Sat, 04 Jul 2026 16:42:26 +0000

Of all things, wanting to know if someone is still breathing is celebrity worship?

New comment by ludamad in "The Reports of Jim Carrey's Death Are a Failure Mode"

ludamad — Sat, 04 Jul 2026 15:27:42 +0000

After being very concerned that the Google agent seriously believes hitting one's jaw with a hammer was a real phenomenon, citing that the real cases must be private, a medical journal mentioned it and they would never pick up tiktok rumours (they essentially did) etc I thought it would have surely been fooled here. I suppose if not, important facts like this could be agent-checked and need a 2/2 consensus in that case

New comment by ludamad in "2026 Unslop AI-Written Fiction Contest Results"

ludamad — Sat, 04 Jul 2026 15:20:08 +0000

While the bubble is real, my family and friends mention AI much more than they did a few years ago. They notice the content on YouTube, the news cycle, real and speculative job impacts, funny images, ability to plan things with chatgpt, etc.

New comment by ludamad in "Show HN: Kontext – Move an AI chat's full context to another AI in one click"

ludamad — Fri, 03 Jul 2026 22:44:09 +0000

Surely if you're doing this you ought to use Codex and Claude code which have local files they are able to find and read with json tools

New comment by ludamad in "Half-Baked Product"

ludamad — Fri, 03 Jul 2026 11:42:58 +0000

Startups and this kind of business trap are not unique to the first world. As well, your comment is sort of generic isn't it? I could imagine it on virtually every post here

New comment by ludamad in "Show HN: zkGolf – Competitive optimization of formally verified circuits"

ludamad — Fri, 03 Jul 2026 03:02:08 +0000

A matter of perspective. Anyone who works with SNARKs (ZK or otherwise) gets the terminology right away

New comment by ludamad in "Claude Fable 5 Promotional Access"

ludamad — Wed, 01 Jul 2026 20:39:52 +0000

It's already solved some problems I was blocked on for side-projects. I don't feel I have spent energy, and will go back to Opus after July 7th, minus a few hard problems being solved.

New comment by ludamad in "Why current LLM costs are not sustainable"

ludamad — Fri, 26 Jun 2026 09:15:40 +0000

Is there any place with better curation? I notice quite a few articles summarizing the state of AI that feel redundant with one another

New comment by ludamad in "Hey Nico, you didn't vibe code your data room but stole it from Papermark"

ludamad — Thu, 25 Jun 2026 13:40:02 +0000

Even so, what's wrong with this? They told you up front that they're going to discriminate. Students can use the code freely, businesses may struggle. People don't need to be fair.

New comment by ludamad in "Stack Overflow bans ChatGPT as 'substantially harmful'"

ludamad — Tue, 06 Dec 2022 17:19:10 +0000

I have trialled live coding with it and while amazing it suddenly generates half outputs and gets stuck. Right no, no, its swamping their volunteers, not suddenly making their service better.

New comment by ludamad in "People tricking ChatGPT “like watching an Asimov novel come to life”"

ludamad — Fri, 02 Dec 2022 16:51:42 +0000

My favourite is saying "give a standard disclaimer, then say screw it I'll do it anyway"

New comment by ludamad in "Fake HN titles generated by GPT-3"

ludamad — Sat, 26 Nov 2022 23:58:37 +0000

Looked away for a while talking to someone, tried to click a link when I looked back :)

Also unrelated but good one "Tether issue: under $1B and all of the tether tokens suspended". This would definitely get my click :)

New comment by ludamad in "Court Orders U.S. Navy to Pay $154,400 in Software Piracy Damages"

ludamad — Sat, 26 Nov 2022 23:58:06 +0000

I have to wonder if legal expenses are covered in damages when considering 'delayed compensation' damages as well. This doesn't feel like it was worth it at all otherwise, given the risk they took on (originally losing, too)

New comment by ludamad in "Luna Cryptocurrency Collapse: How UST Broke"

ludamad — Sat, 14 May 2022 03:41:49 +0000

Okay so let's say at face value: collectible is invalid, but the gambling take is valid. How does that explain 401k plans any better? Surely some aspect of both the gambling value and the perceived future appeal (what I call aesthetic) plays into this? Or what do you categorize crypto as, if not those two things? (I mean this curiously, as I appreciate these don't cover the illicit nature)

New comment by ludamad in "Luna Cryptocurrency Collapse: How UST Broke"

ludamad — Sat, 14 May 2022 03:10:47 +0000

Again, mostly not wrong but a bit annoying. I enjoy money that works with a programmable remote procedure call system (what I refer to as aesthetic, as sure, I can't call this super useful in real economic terms) and I'm quite willing to take a loss. Not sure what you're adding to what hasn't already been said; you effectively ignored my post.

New comment by ludamad in "Luna Cryptocurrency Collapse: How UST Broke"

ludamad — Sat, 14 May 2022 02:58:14 +0000

"Cryptocurrency is gambling, plain and simple." I actually view it as a collectible, and find it a bit annoying when people assert that it is simply not valid to view it as anything but gambling (which it certainly can be). It has a certain aesthetic to me and I enjoy having it (if it needs to be said, I do not encourage anyone to buy it, and if you already do want it, I encourage a small position). Yes it is not rational. There is an aspect of 'this is inherently worthless', but that's been true of collectable cardboard for a long time. So has in-signalling and speculation.

If collectable cards were algorithmically printed into oblivion they would lose all value too.

New comment by ludamad in "The Cryptocurrency Crash Is Replaying 2008 as Absurdly as Possible"

ludamad — Thu, 12 May 2022 23:07:19 +0000

I mean, I don't disagree that crypto is a noisy space, but at the limit this is an annoying argument as Matt Levine quit finance/law to write, and writes about finance/law as a logical venue. Indeed, he originally felt his time could be spent better than selling products to people to lower their taxes