Hacker News: zerd

New comment by zerd in "DeepSeek-v4.1-Flash-1017"

zerd — Sun, 18 Oct 2026 04:39:19 +0000

Interesting, I made a Claude code plugin to do just that two weeks ago.

New comment by zerd in "DeepSeek-v4.1-Flash-1017"

zerd — Sun, 18 Oct 2026 04:38:22 +0000

With a subscription the main conversation has a 1h TTL, and subagents 5m. https://code.claude.com/docs/en/prompt-caching#cache-lifetim...

New comment by zerd in "When did Google get so weird?"

zerd — Mon, 28 Sep 2026 21:28:53 +0000

Claude just used the word “ Netowrk” instead of Network. Mind you, I was only using Sonnet with reasoning, but I thought trillion-parameter models would be able to spell a common word.

New comment by zerd in "When did Google get so weird?"

zerd — Mon, 28 Sep 2026 20:43:29 +0000

Gemini is the worst gaslighter. It hallucinated a feature in an open source product. I said no that does not exist. It invented very realistic git commit messages. I said commit does not exist, so it said I was out of sync. I linked the repo and it said it might be a fork. Then it invented a mailing list thread where they supposedly removed the feature, that’s why I couldn’t find it.

New comment by zerd in "When did Google get so weird?"

zerd — Mon, 28 Sep 2026 20:22:00 +0000

Searching for anything product related is now AI answer, followed by “Sponsored products”, followed by “Popular products”, a few hits, then “Shop by store”, and “In store nearby”. Maybe 2% is actual content, the rest is ads.

Their prediction was spot on: “We expect that advertising-funded search engines will be inherently biased toward the advertisers and away from the needs of the consumers.”

New comment by zerd in "AX – Google’s Open Agentic Orchestrator"

zerd — Mon, 21 Sep 2026 17:10:47 +0000

Limiting it to just read and http limits its usefulness. If you want something like “filter for this, count the number of matches, format like this” you have to make custom tools. And you have to make sure they don’t have any bugs that allow arbitrary code execution. You’re effectively building your own sandbox in this case. Using a standard one sounds easier unless you have a very focused use case.

New comment by zerd in "Three sites made 215,128 “best software” pages for AI. Perplexity cites them"

zerd — Sat, 05 Sep 2026 23:56:59 +0000

Time to start some new restaurants matching those. Like Bubba Gump.

New comment by zerd in "Samsung's Processing-in-Memory (PIM)"

zerd — Sun, 30 Aug 2026 05:04:09 +0000

How do you scale past 16 channels though? 16 channels give you around 614 GB/s, while PIM can do that per chip, so it can achieve 58TB/s.

New comment by zerd in "Mythic's analog compute-in-memory architecture"

zerd — Fri, 28 Aug 2026 04:57:12 +0000

Veritasium did a video on this 4 years ago. https://www.youtube.com/watch?v=GVsUOuSjvcg

New comment by zerd in "GLM-5.3-Flash"

zerd — Wed, 26 Aug 2026 18:29:41 +0000

I’ve been doing the same thing, giving the same tasks to Qwen 3.8 27B and Opus, and the main difference is that Qwen does not consider edge cases which Opus catches. It’s good at the happy path, but even when hinting that there are uncovered edge cases and gotchas it’s oblivious to it. So I feel like I need a bigger model to do planning/review.

New comment by zerd in "Google Chrome update will close the door on ad blockers"

zerd — Tue, 16 Jun 2026 18:43:36 +0000

"We're doing it for your own good, pinky promise"

New comment by zerd in "FastContext: Training Efficient Repository Explorer for Coding Agents"

zerd — Tue, 16 Jun 2026 17:41:41 +0000

I noticed Claude code does something similar when "Exploring" the code base, it spawns a subagent using Heroku to get the summary and file locations, which is both faster and it doesn't pollute the context as much.

I wonder how it compares to an vector indexing approach.

New comment by zerd in "Running local models is good now"

zerd — Tue, 16 Jun 2026 17:00:16 +0000

I'd love an RTX 6000 Pro, but how can you justify it when it costs 10 years worth of Claude Max?

New comment by zerd in "Running local models is good now"

zerd — Tue, 16 Jun 2026 16:43:08 +0000

I noticed Fable was quite a bit terser, and I think it's due to changes in the system prompt [0]. They're literally saying "just give me the TLDR" and "give brief updates". You can tweak a lot of that with an AGENTS.md.

[0] https://twelvetables.blog/comparing-claude-fable-5s-system-p...

New comment by zerd in "Gmail thinks I'm stupid, so I left"

zerd — Wed, 03 Jun 2026 01:57:17 +0000

Thunderbird does not seem to have auto-categorization from what I see, just filtering. Neither does KMail. Unless you’re referring to some addons? For apple mail you have to add it on each client. And a lot of comments are about how to disable it because it categorizes wrong.

New comment by zerd in "YouTube to automatically label AI-generated videos"

zerd — Fri, 29 May 2026 03:42:15 +0000

Reddit “best” sorting is pretty much like instagram and TikTok now, have to make sure it on hot/top, otherwise it’ll show you “related” things from subreddits you never subscribed to.

New comment by zerd in "Issue links now open in a popup"

zerd — Mon, 27 Apr 2026 16:35:01 +0000

Also it breaks copying links. If I want to link to an issue I copy the URL. But now there's two different issues open at the same time, which one am I linking to? Original? Popup? Both?

New comment by zerd in "Voyager 1 runs on 69 KB of memory and an 8-track tape recorder"

zerd — Sun, 29 Mar 2026 22:17:07 +0000

Based on the communication fix, they also didn't have a simulator, or tests, or complete source code, on a custom instruction set that wasn't well documented, so they had to reverse engineer how it worked. https://www.youtube.com/watch?v=YcUycQoz0zg&t=2366s

New comment by zerd in "Nanobrew: The fastest macOS package manager compatible with brew"

zerd — Wed, 25 Mar 2026 04:05:36 +0000

It's ok because I rarely need it, but sometimes it's so slow I'm looking for alternative install methods while it's doing it's thing.

New comment by zerd in "Push events into a running session with channels"

zerd — Fri, 20 Mar 2026 01:11:49 +0000

I was making a telegram to Claude via tmux capture-pane and send-keys, this will be so much nicer. Also sounds like something that addresses some of what Steve Yegge said was missing for agent to agent communication as well.