Hacker News: gardnr

New comment by gardnr in "Show HN: Vegetable Gardening for Engineers"

gardnr — Sun, 08 Nov 2026 17:48:03 +0000

When people discover that I work in tech they ask me what they should buy. I tell them that they should buy seeds.

New comment by gardnr in "Tell HN: Gemini trains on chat history, even for subscribers (personal accounts)"

gardnr — Wed, 04 Nov 2026 14:50:10 +0000

Oh, I get it now. The anti-user behaviour is actually there to help me as a user. The main goal isn’t to Hoover up all my data to profile me and serve more targeted ads in more places. The main goal is to make search better. Now I get it.

New comment by gardnr in "GitHub's new dashboard experience now the default"

gardnr — Sat, 03 Oct 2026 11:00:26 +0000

Over the last four years, GitHub announcing a change or new feature has generally led to a worse experience, for me at least.

New comment by gardnr in "Qwen 3.8 27B available on Cerebras at 1500 tokens/s"

gardnr — Thu, 03 Sep 2026 19:20:31 +0000

Last time I got one, I had to log into a Discord server and wait for "the drop" and IIRC Daniel Kim was giving them out based on who was there at the time. They were gone in less than a minute. This was ~8 months ago.

New comment by gardnr in "Qwen 3.8 27B available on Cerebras at 1500 tokens/s"

gardnr — Thu, 03 Sep 2026 18:44:51 +0000

They make a giant inference chip. Their inference service is basically just advertising for their core value prop: hardware.

The CEO was on Gradient Dissent a couple years ago: https://www.youtube.com/watch?v=qNXebAQ6igs

New comment by gardnr in "Qwen 3.8 27B available on Cerebras at 1500 tokens/s"

gardnr — Thu, 03 Sep 2026 18:37:49 +0000

I used their Coding Plan for a few months. It is genuinely difficult to keep up with the models. The output is so fast. Qwen 3.8 27B is likely one of the strongest models they've hosted so far.

Edit: it looks like this is only available on a API token pricing. Does anyone know if they have rolled out prompt caching yet? It used to get pretty expensive for agentic coding tasks with no prompt caching.

New comment by gardnr in "Google Has Removed MV2 Extensions from the Chrome Web Store, Including UBO"

gardnr — Mon, 31 Aug 2026 21:36:05 +0000

I remember in 2010 when Chrome made the web so much better for everyone. All the early adopters were singing its praise and encouraging friends and family to use it.

Now, I encourage people to use anything but Chrome. Firefox is really great these days.

New comment by gardnr in "Norway Shrugged (2024)"

gardnr — Sun, 30 Aug 2026 20:59:56 +0000

This article is from 2024. They can defer this tax now. [1]

> High income with short work days, free healthcare, free daycare, free education and beyond.

If the author doesn't value these for the people around them then perhaps he should move somewhere else.

1: https://www.skatteetaten.no/en/person/taxes/tax-deduction-ca...

New comment by gardnr in "Harness Engineering"

gardnr — Thu, 27 Aug 2026 17:15:50 +0000

Interesting: I initially confused smol for https://github.com/huggingface/smolagents

New comment by gardnr in "Show HN: RealDiff – runtime behavior diffing for pull requests (six languages)"

gardnr — Thu, 27 Aug 2026 15:39:56 +0000

Don’t run your responses through an LLM. People prefer to read your genuine words, even if it may include some incorrect English.

New comment by gardnr in "Harness Engineering"

gardnr — Thu, 27 Aug 2026 15:07:48 +0000

That looks like a “Deep Research” report generated by an LLM, not a summary.

New comment by gardnr in "Harness Engineering"

gardnr — Thu, 27 Aug 2026 14:40:07 +0000

Here’s the original article this one is based on: https://martinfowler.com/articles/harness-engineering.html

And here is the original author on a podcast talking about it: https://open.spotify.com/episode/4FxEdjXldNhoh67KYVmbDu

New comment by gardnr in "RAG Is Simpler Than You Think"

gardnr — Wed, 26 Aug 2026 15:49:47 +0000

Nice work.

New comment by gardnr in "RAG Is Simpler Than You Think"

gardnr — Wed, 26 Aug 2026 15:40:00 +0000

That is the approach I would take today. Late Interaction is worth a look. Evals are necessary.

New comment by gardnr in "London Underground begins scanning passengers' faces"

gardnr — Wed, 12 Aug 2026 10:37:33 +0000

Wash your hands? /s

New comment by gardnr in "Tax cuts for the wealthy only benefit the rich (2023)"

gardnr — Fri, 07 Aug 2026 17:36:08 +0000

I want a system that taxes things we don't want (unproductive rent-seeking) and doesn't tax things we do want (productive stuff e.g. income tax).

I don't want to be personally compensated. I just want to live in a society where people's basic needs are easily met: clothes, food, medicine, housing. The current landscape is, by the day, increasingly dystopian.

New comment by gardnr in "Qwen 3.8"

gardnr — Sun, 19 Jul 2026 09:08:27 +0000

It's hard to say what their motivation is. The Chinese firms seem to be working hard to commoditize intelligence which may be the most effective way to debase American frontier labs. And yeah: it also happens to be really good for humanity.

New comment by gardnr in "[dead]"

gardnr — Mon, 13 Jul 2026 21:42:19 +0000

No GH contributors. New HN account. Vibe coded in 5 commits over the last two week.

Interesting choice of model:

> Co-Authored-By: Claude Opus 4.5

> 2 weeks ago

---

Opus 4.5 was introduced in November of 2025 (8 months ago)

Opus 4.6 was introduced in February of 2026 (5 months ago)

Opus 4.7 was introduced in April of 2026 (3 months ago)

Opus 4.8 was introduced in May of 2026 (2 months ago)

From: https://github.com/Planwright/planwright/commit/c746cdcaf7e6...

New comment by gardnr in "Samsung Health app threatens data deletion if users opt out AI training"

gardnr — Mon, 13 Jul 2026 21:32:15 +0000

This is like Google Ultra for personal accounts. I signed up to see what it was like and then assumed I would be able to disable training on my data as a paid customer. The only way to disable training on paid personal accounts is to disable history (no chat logs) which makes the service much less useful for me.

For Google Workspace accounts that use the Ultra plan you can disable training while retaining history. I didn't bother signing up again. It is user-hostile.

New comment by gardnr in "LARP – Revenue infrastructure for serious founders"

gardnr — Sun, 12 Jul 2026 21:01:05 +0000

I too was nervous until I clicked through and had a read.