Hacker News: nialse

New comment by nialse in "Gemma 4 12B: A unified, encoder-free multimodal model"

nialse — Thu, 04 Jun 2026 17:23:29 +0000

The FFT is essentially just a matrix multiplication, or two. No need for fancy conversions. Just a huge amount of training data and a very large array.

New comment by nialse in "Texas Adds Another Solar Farm as Ercot Grid Demand Soars"

nialse — Wed, 03 Jun 2026 15:14:05 +0000

The power needs of a house of course vary with many parameters. Our 100 square meter house in Nordic climate draws 1.2 kW on average over a year including charging the car. What are common numbers elsewhere?

New comment by nialse in "Codex just found a "workaround" of not having sudo on my PC"

nialse — Sun, 31 May 2026 21:15:54 +0000

Something like that.

New comment by nialse in "Codex just found a "workaround" of not having sudo on my PC"

nialse — Sun, 31 May 2026 19:36:33 +0000

This was of course dependent on yolo mode, but automatic approval has also been pulling stunts like this. A recent example is data that was purposely kept away from Codex in a folder far far away. When it found a single reference it just went for the data when having an issue. Lesson learned, keep essential data and Codex separated on different machines. Codex remote ssh actually helps here.

New comment by nialse in "Hallucinate – Massively Multiplayer Online Rave"

nialse — Thu, 28 May 2026 07:37:41 +0000

The closest we've ever been!

New comment by nialse in "I believe there are entire companies right now under AI psychosis"

nialse — Sat, 16 May 2026 05:33:18 +0000

”Religious suffering is, at one and the same time, the expression of real suffering and a protest against real suffering.

Religion is the sigh of the oppressed creature, the heart of a heartless world, and the soul of soulless conditions.

It is the opium of the people.”

Some are on copium, some on hopium. The gods change names; the need for relief remains.

New comment by nialse in "I believe there are entire companies right now under AI psychosis"

nialse — Sat, 16 May 2026 05:15:49 +0000

Keen observation. Maybe automation will come for the AI as well?

New comment by nialse in "I believe there are entire companies right now under AI psychosis"

nialse — Sat, 16 May 2026 05:13:05 +0000

The people were longing for utopia, just not the same utopia.

New comment by nialse in "Mitchellh – I strongly believe there are entire companies now under AI psychosis"

nialse — Fri, 15 May 2026 21:18:10 +0000

Another argument for less human-like AI then, I guess.

New comment by nialse in "I believe there are entire companies right now under AI psychosis"

nialse — Fri, 15 May 2026 21:05:37 +0000

I'm starting to long for the age after AI. When the generative euphoria has settled and all outputs are formally verified based on exquisite architectures and standards.

Thoughts on Doctoral Student Recruitment

nialse — Tue, 28 Apr 2026 09:59:49 +0000

Article URL: https://www.youtube.com/watch?v=8UNLwpwGnb0

Comments URL: https://news.ycombinator.com/item?id=47932391

Points: 2

# Comments: 0

Automated systematic literature review with Claude Code

nialse — Sun, 26 Apr 2026 19:31:55 +0000

Article URL: https://www.youtube.com/watch?v=1K_4QqUlSBU

Comments URL: https://news.ycombinator.com/item?id=47913221

Points: 2

# Comments: 0

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 11:03:49 +0000

I see your point. Many of my prompts for reasoning ends with: No code. Planning mode is sort of the workaround for this specific situation. Sometimes it is useful for the AI agent just to think. It looks like I need a screwdriver in addition to the aforementioned hammer, a pozidriv screwdriver to be precise.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 10:29:39 +0000

To be fair, there is likely not much training data on the difficult conversations you need to handle in a senior position, pushback being one of them. The trouble for the agents is that it is post hoc, to explain themselves, rationalising rather than ”help me understand” beforehand.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 09:54:49 +0000

I think that it is a fair perspective to allow role play, and it's useful too, when explicit. Does not really make sense for AI to cosplay human all the time though.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 09:53:02 +0000

Maybe we should change that? Of course symbolic AI was the holy grail until statistical AI came in and swept the floor. Maybe something else though.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 08:51:13 +0000

The other day Codex on Mac gained the ability to control the UI. Will it close itself if instructed though? Maybe test that and make a benchmark. Closebench.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 08:48:52 +0000

Asking for code to manipulate the AST is another route. In python it can do absolute magic.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 08:38:00 +0000

Interesting. When I code, I want a boring tool that just does the work. A hammer. I think we agree on that the tool should complete the assignment reliably, without skipping parts or turning an entirely implementable task into a discussion though.

New comment by nialse in "Less human AI agents, please"

nialse — Tue, 21 Apr 2026 08:27:39 +0000

That is probably the next step, and in practice it is much of what sub-agents already provide: a kind of tabula rasa. Context is not always an advantage. Sometimes it becomes the problem.

In long editing sessions with multiple iterations, the context can accumulate stale information, and that actively hurts model performance. Compaction is one way to deal with that. It strips out material that should be re-read from disk instead of being carried forward.

A concrete example is iterative file editing with Codex. I rewrite parts of a file so they actually work and match the project’s style. Then Codex changes the code back to the version still sitting in its context. It does not stop to consider that, if an external edit was made, that edit is probably important.