Hacker News: maCDzP

New comment by maCDzP in "Douglas Hofstadter: Analogy as the Core of Cognition [video]"

maCDzP — Thu, 10 Sep 2026 19:50:00 +0000

I lover his book ”I am a strange loop”. It has shaped how I view myself.

New comment by maCDzP in "An Accidental Blackboard"

maCDzP — Wed, 09 Sep 2026 03:36:24 +0000

I found this really interesting. You could probably prompt it to use a ”blackboard system” and reference the wiki. Maybe a place to store the data is as a GitHub discussion/issue or something similar. That way you can easily browse it and watch what’s happening.

New comment by maCDzP in "Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses"

maCDzP — Tue, 08 Sep 2026 17:43:08 +0000

Thank you for these, coz I learned a lot! Great that they are open access.

New comment by maCDzP in "Three important steps in my maturation process"

maCDzP — Sat, 22 Aug 2026 08:02:55 +0000

I agree fully with the point about integrating emotion in your decision making. Earlier I used to get stuck in decisions since I had the need to be rational. Which is in it self an emotion I guess - since need is an emotion? Now my life is much simpler.

New comment by maCDzP in "Maximizing the value of your Claude Code sessions"

maCDzP — Sat, 15 Aug 2026 09:11:09 +0000

I have had some success with using Ollama cloud and just instructing Claude Code to hand off tasks to Ollama because of tokens economics.

New comment by maCDzP in "I added a real-time chat to my blog, people used it to attack me"

maCDzP — Thu, 06 Aug 2026 15:40:00 +0000

If the chat is saved as an example of how lot to talk. Can that be used for training an LLM on how to talk or what to remove from a chat?

New comment by maCDzP in "Cloudflare OS: an open platform for agents, apps, and work"

maCDzP — Thu, 06 Aug 2026 11:20:32 +0000

Reminds me of EMacs, which I really like.

New comment by maCDzP in "Discovery Loop"

maCDzP — Thu, 06 Aug 2026 11:18:07 +0000

Heh, I run my experiments on Claude subscription. And that cost I’ll file as an ”opportunity cost”. But if I would have build this by hand and given myself a salary, then paying Claude to do it is way cheaper.

New comment by maCDzP in "Discovery Loop"

maCDzP — Wed, 05 Aug 2026 20:37:14 +0000

I have used something similar. I set up a team of agents that researches, proposes, builds and audits. Then rinse and repeat. I have used it for different topics. It hasn’t made me a millionaire, but I haven’t lost money either - so that’s some sort of win, right? But I would not have been able to ideate, test at that speed and quality without an LLM.

New comment by maCDzP in "Now is the time to give LLMs access to the ACM digital library"

maCDzP — Tue, 28 Jul 2026 20:17:51 +0000

If it was a non profit that trained the model - would that change your mind?

New comment by maCDzP in "Opus 5 is currently #1 on Artificial Analysis Intelligence Leaderboard"

maCDzP — Sat, 25 Jul 2026 06:15:44 +0000

I have had success with ”brainwashing” by starting out with bug bounties/CTF and then going from there.

New comment by maCDzP in "Claude Opus 5"

maCDzP — Fri, 24 Jul 2026 19:20:53 +0000

Has anyone tried that? I have a feeling that if I put it a prompt Claude would comply. But I am all in on the Claude cool aid.

New comment by maCDzP in "All 253 Patterns from Christopher Alexander's a Pattern Language Summarized"

maCDzP — Thu, 23 Jul 2026 07:17:07 +0000

I work in civil engineering and tinker with computers as a hobby. I found Alexander’s work delightful. It really helped me think about my profession and my hobby and how they tie together.

He also talks about community in a city, so it walks across all scales in the build world. That really helped me think about ”scale” and what phenomena emerges at different scales. I haven’t seen those thoughts about community when I have read about patterns in SWE.

Patterns mostly revolves around solving specific problem in programming, but I haven’t yet found something that uses the same thoughts on a larger scale explicitly. I believe there are patterns for how to architect multiple softwares, but what’s beyond that scale? Say how to integrate different SWE communities.

New comment by maCDzP in "Your code is fast – if you're lucky"

maCDzP — Sat, 11 Jul 2026 13:51:56 +0000

I did Nand2Tetris and then I understod why I need to vectorize. You get a view from nand-gate to software and get to see all the interfaces. Its a wonderful course.

New comment by maCDzP in "Professor denounces mass AI fraud on an exam at Brown"

maCDzP — Mon, 29 Jun 2026 03:56:39 +0000

During uni I only took a Java class. During the exam I had to turn in hand written code. I guess that would work today.

New comment by maCDzP in "Ask HN: What was your "oh shit" moment with GenAI?"

maCDzP — Sat, 06 Jun 2026 18:42:33 +0000

This. I managed to run Gemma 4 31B on AMD MI250X for CPT and SFT with Claude. And I have ZERO experience and knowledge of how to train and work with GPU:s. The training didn’t go where I wanted. But I manage to direct the AI to build it. It’s crazy. I am excited and scared.

New comment by maCDzP in "Show HN: Boxes.dev: ditch localhost; run Claude Code and Codex in the cloud"

maCDzP — Thu, 04 Jun 2026 18:16:48 +0000

I run Claude Code on my VPS and do /rc to run from my mobile. It’s really handy.

New comment by maCDzP in "Bringing Up DeepSeek-V4-Flash on AMD MI300X"

maCDzP — Tue, 02 Jun 2026 20:49:08 +0000

I train on AMD MI250X and managed to get Gemma 4 31B to work - but it took a lot of work on the software side.

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

maCDzP — Sun, 10 May 2026 04:03:10 +0000

I have tried that and it works well for me. I can also recommend using a Visual Studio plugin called Markco that lets you comment a markdown file. It’s resembles words commenting feature. I use it to comment LLM generated markdown.

New comment by maCDzP in "Research-Driven Agents: When an agent reads before it codes"

maCDzP — Thu, 09 Apr 2026 19:49:13 +0000

I have a ML project. I usually set up a team of agents, where I have a leader, archivist, research assistant, researcher, developer and tester. The team generates hypothesis based on papers, test it, and iterate over that. Everything is documented using a lab notebook. It burns tokens but I have found some promising strategies that I am testing.