Hacker News: kosolam

New comment by kosolam in "Show HN: I canceled my AI code reviewer and wrote a free local one"

kosolam — Tue, 18 Aug 2026 13:49:29 +0000

Can you share some highlights of what it can do? (I’m just asking the agent to review the pr usually)

New comment by kosolam in "Harness engineering for self-improvement"

kosolam — Tue, 04 Aug 2026 21:39:34 +0000

What do you mean please give more information.

New comment by kosolam in "Frame – Linux X server in Assembly"

kosolam — Fri, 17 Jul 2026 20:05:54 +0000

״I am not sure this laptop has a fan anymore. Except me״ huh nice one

New comment by kosolam in "Show HN: Juggler – an open-source GUI coding agent, by the creator of JUCE"

kosolam — Tue, 14 Jul 2026 18:54:00 +0000

Lol

New comment by kosolam in "Codex starts encrypting sub-agent prompts"

kosolam — Tue, 14 Jul 2026 12:31:35 +0000

But codex is opensource, no?

New comment by kosolam in "Show HN: A statically typed, cross-platform, easily bootstrappable build system"

kosolam — Sat, 04 Jul 2026 08:01:17 +0000

Looks like here lua vm is used as the basis for the project itself not as the language u write your builds with

New comment by kosolam in "Ask HN: Is anyone experimenting with different ways of using LLMs for coding?"

kosolam — Fri, 03 Jul 2026 21:24:29 +0000

Doing similar stuff. I curse a lot. The processes do transfer mostly between different models which a bit validates the approach.

New comment by kosolam in "Ask HN: Is anyone experimenting with different ways of using LLMs for coding?"

kosolam — Fri, 03 Jul 2026 21:18:34 +0000

I never looked back. Full control, and it’s really basic stuff you can have in an hour or so. And then you stop fighting with the external dependency and can move fast with your ideas.

New comment by kosolam in "Ask HN: Is anyone experimenting with different ways of using LLMs for coding?"

kosolam — Fri, 03 Jul 2026 20:06:55 +0000

Ha I’m doing the same. 2 months in or so, lost track of time a bit. Started with running all the usual suspects pi, claude code, codex, goose, but mostly migrated to own agent (ha!) for greater level of control. Also doing it in python for C lang as first target (ha!). Anyhow post a link if you’d release yours I’d be interested to take a look at adjacent work.

New comment by kosolam in "IBM debuts sub-1 nanometer chip technology"

kosolam — Fri, 26 Jun 2026 09:27:33 +0000

They dont give the exact size hmm

New comment by kosolam in "MAI-Code-1-Flash"

kosolam — Wed, 03 Jun 2026 10:37:42 +0000

Hey Dave, I’d love to add your new model in the harness I’m going to opensource very soonish. Going to publish benchmarks on real world tasks.

New comment by kosolam in "Why AI Makes Us So Angry"

kosolam — Tue, 02 Jun 2026 10:29:22 +0000

Yeah, pretty much aligns with my own observations… I think the rage attacks began when Opus 4.7 was released more or less kah kha

Ask HN: Where AI Researchers Congregate?

kosolam — Sat, 23 May 2026 10:05:47 +0000

So I’m doing plenty of experiments and applied research in autonomous agents and agentic flows in general. I’m looking for a place where I could collaborate and discuss with other like minded people. I would love to get recommendations for such a place online.

Comments URL: https://news.ycombinator.com/item?id=48246397

Points: 2

# Comments: 0

New comment by kosolam in "Show HN: Rmux – A programmable terminal multiplexer with a Playwright-style SDK"

kosolam — Thu, 21 May 2026 10:50:02 +0000

But wait, I’m a user of tmux or zellij which for me was basically another tmux and I went back to tmux eventually. I’m not looking to develop another tmux or something. Is this project more of a library to create tmux like apps?

New comment by kosolam in "Show HN: Rmux – A programmable terminal multiplexer with a Playwright-style SDK"

kosolam — Thu, 21 May 2026 09:56:12 +0000

Hey, I automate tmux all the time with send keys and capture.. how is your project improves vs tmux or zellij?

New comment by kosolam in "Show HN: Forge – Guardrails take an 8B model from 53% to 99% on agentic tasks"

kosolam — Wed, 20 May 2026 09:36:32 +0000

Hey, this is most probably related to the chat template or the reasoning parser or the tool call parser or also things like kv cache quantization and possibly other params that affect results like the regular top k top p and all of that, the backend often sets its own defaults or the lack of them. It’s best to have all these under control if possible. I wonder regarding this project have you been testing it on real world projects? I’m working on an agentic loop as well also using a local model.

New comment by kosolam in "Library for fast mapping of Java records to native memory"

kosolam — Mon, 11 May 2026 20:04:04 +0000

Nice. Very clean api.

New comment by kosolam in "Ask HN: Did you change the site on mobile?"

kosolam — Fri, 06 Mar 2026 16:52:20 +0000

For anyone that ever reads this, the issue was associated with the zoom level, up to a certain zoom level it behaves ok and wraps the text.

New comment by kosolam in "Ask HN: Did you change the site on mobile?"

kosolam — Thu, 05 Mar 2026 22:15:17 +0000

Worked fine for many years for me. Not great, but usable.

New comment by kosolam in "Ask HN: Did you change the site on mobile?"

kosolam — Thu, 05 Mar 2026 22:14:19 +0000

I mean in portrait. In landscape it’s fine.