Hacker News: CjHuber

New comment by CjHuber in "Zero Knowledge Tolstoyan Art"

CjHuber — Tue, 14 Jul 2026 13:04:14 +0000

Every good thing is alike, every slop is sloppy in its own way

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

CjHuber — Tue, 14 Jul 2026 11:43:28 +0000

Given that codex itself now ships a proxy that wraps the subscription, it seems unlikely

https://github.com/openai/codex/blob/main/codex-rs/responses...

New comment by CjHuber in "Claude Code sends 33k tokens before reading the prompt; OpenCode sends 7k"

CjHuber — Sun, 12 Jul 2026 20:00:44 +0000

> Probably because the general purpose subagents inherit the parent model

only if you don't specify which model should be used

OpenAI's API can now keep reasoning across turns instead of discarding it

CjHuber — Fri, 10 Jul 2026 02:19:07 +0000

Article URL: https://drop-05a4352b-803.sophisticated-stay.workers.dev

Comments URL: https://news.ycombinator.com/item?id=48854997

Points: 3

# Comments: 0

New comment by CjHuber in "GPT-5.6"

CjHuber — Thu, 09 Jul 2026 17:41:43 +0000

> Instead of requiring developers to script every step or passing every tool response back through the model, Programmatic Tool Calling in the Responses API can filter large amounts of intermediate data, retain only what matters, and adapt its workflow along the way.

this seems very interesting

New comment by CjHuber in "GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday"

CjHuber — Wed, 08 Jul 2026 10:29:35 +0000

but 5.5 was also no major version bump

New comment by CjHuber in "GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday"

CjHuber — Wed, 08 Jul 2026 08:16:05 +0000

> fit a lot more work inside the model's context window before you're forced into a slow and lossy compaction step

This is something I never understood. Why the reasoning is not included until the context is full, then the reasoning stripped optionally to allow the conversation to continue. and only then when its truly full offer a compaction. Was it to optimize caching? Well I guess it doesn't matter now that you hinted that this choice was made because of prior limitations and may change very soon

New comment by CjHuber in "GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday"

CjHuber — Wed, 08 Jul 2026 07:53:13 +0000

Well, obviously its true as you can see in the silbing comment. But still interesting how experiences differ. I wonder did you ever use codex in more demanding workflows, because that at least something like that is going on is obvious if you use it for a few hours

New comment by CjHuber in "GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday"

CjHuber — Wed, 08 Jul 2026 05:25:25 +0000

Interesting to hear people like gpt-5.5. For me it feels smart only at one shot prompts, but if you try to build up session context before doing something it feels magnitudes inferior to Claude. I'm almost sure its because the thinking of previous turns is stripped with the responses API, so if I tell it to analyse something deeply, what remains of the understanding in future turns is only the short response text of that analysis

New comment by CjHuber in "The family keeping watch over a 52-year-old pot of soup"

CjHuber — Wed, 08 Jul 2026 03:55:46 +0000

> so you now have to wait in line for hours

Interesting, I go there pretty regularly and while It is well visited, I not once encountered a queue there

New comment by CjHuber in "GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance"

CjHuber — Sun, 05 Jul 2026 16:20:59 +0000

I agree, adaptive thinking is a pest and without a minimum thinking budget especially Claude for me currently defaults to not think at all even on max effort.

Sequential-thinking was really a step in the right direction, and works almost exactly how you've described, though when it was popular before the reasoning models and even now when I tried it recently I have never once see it use its branching feature and it tends also to have the RLHF urge to answer something "helpful" quickly instead.

New comment by CjHuber in "Show HN: Hackers for Granny (defense against industrialized elder fraud)"

CjHuber — Tue, 16 Jun 2026 15:04:54 +0000

I think this is a really good cause but I can't help but wonder about the contrast between the puppy analogy and then increasing the friction of installing it by making it a paid tool

New comment by CjHuber in "AI Has Amnesia. Here's Every System Built to Fix It"

CjHuber — Mon, 15 Jun 2026 08:01:50 +0000

What about GBrain

New comment by CjHuber in "Don't trust large context windows"

CjHuber — Sun, 14 Jun 2026 10:41:57 +0000

Yeah indeed it's very interesting. And the 60k initial context don't even contain the suggested change yet. For me if I don't do this the current models tend to fixate and local patches instead of tracing symbols and making a holistic model of what a change interacts with in the codebase

New comment by CjHuber in "Don't trust large context windows"

CjHuber — Sun, 14 Jun 2026 08:44:39 +0000

I'm always a bit confused when people say things like this. 60k token is often more than the initial context I feed the model with. And I don't think I ever had a productive session that began under 150k tokens.

New comment by CjHuber in "macOS 27 Beta breaks the ability to boot Asahi Linux"

CjHuber — Thu, 11 Jun 2026 18:18:34 +0000

So isn't that just purely security by obscurity then? Would they not rather have someone publicly break it instead of selling a zero day?

New comment by CjHuber in "Google Chrome is killing all uBlock Origin bypasses, Edge, Opera to follow"

CjHuber — Wed, 10 Jun 2026 07:50:13 +0000

Technically yes

New comment by CjHuber in "Nvidia Surpasses Germany"

CjHuber — Sun, 17 May 2026 12:29:43 +0000

I‘m glad you didn’t write value

New comment by CjHuber in "Ghostty is leaving GitHub"

CjHuber — Wed, 29 Apr 2026 07:10:04 +0000

Are you… explaining the effects of being acquired to Hashimoto?

New comment by CjHuber in "An update on recent Claude Code quality reports"

CjHuber — Fri, 24 Apr 2026 01:37:22 +0000

It does have an built in documentation subagent it can invoke but that doesn’t help much if they don’t document their shenanigans