Hacker News: jodleif

New comment by jodleif in "Qwen 3.8 27B"

jodleif — Sat, 15 Aug 2026 05:55:19 +0000

Then you might be missing SWA. Gemma models are extremely memory hungry without

New comment by jodleif in "The worthlessness of Vitamin D is mildly exaggerated"

jodleif — Wed, 24 Jun 2026 14:49:09 +0000

Its about the angle yes, but the main thing is basically for every square unit of solar radiation gets projected onto a larger surface because of the angle- meaning radiation per unit area is lower. The secondary effect is that sunlight has to travel through more atmosphere, and gets scattered somewhat.

New comment by jodleif in "Removable batteries in smartphones will be mandatory in the EU starting in 2027"

jodleif — Mon, 04 May 2026 16:33:36 +0000

Would be really nice. Seems like even android is getting more and more locked down

New comment by jodleif in "Make macOS consistently bad unironically"

jodleif — Sat, 28 Mar 2026 08:27:52 +0000

Yes, except for the odd anti-cheat that needs kernel level access

New comment by jodleif in "AMD's Ryzen 9 9950X3D2 Dual Edition crams 208MB of cache into a single chip"

jodleif — Sat, 28 Mar 2026 08:25:49 +0000

Threadripper is a good alternative. No point having a lot of dual channel ram for LLMs, too slow

New comment by jodleif in "Peter Thiel's 'Steroid Olympics' Startup Wants to Sell You the Sketchy Peptides"

jodleif — Sat, 21 Mar 2026 09:17:19 +0000

In any functioning sports body testing is random and unannounced

New comment by jodleif in "The 100 hour gap between a vibecoded prototype and a working product"

jodleif — Sun, 15 Mar 2026 15:27:12 +0000

As long as you tune the JVM right it can be faster. But its a big if with the tune, and you need to write performant code

New comment by jodleif in "Coding after coders: The end of computer programming as we know it?"

jodleif — Sat, 14 Mar 2026 18:57:12 +0000

Agreed. Also, it takes time to understand a domain properly- so the innate slowness of coding helps with letting things “simmer” in the back of the mind.

New comment by jodleif in "Zig – Type Resolution Redesign and Language Changes"

jodleif — Wed, 11 Mar 2026 12:27:10 +0000

I found swift way more enjoyable than rust as a C++ alternative. It even has first class-ish interop now.

New comment by jodleif in "UUID package coming to Go standard library"

jodleif — Sat, 07 Mar 2026 10:44:03 +0000

Deterministic uuids is a very standard usecase

New comment by jodleif in "Tell HN: I'm 60 years old. Claude Code has re-ignited a passion"

jodleif — Sat, 07 Mar 2026 10:39:42 +0000

The sad part is the “buddy hackathon” is kind of redundant now

New comment by jodleif in "How to code Claude Code in 200 lines of code"

jodleif — Fri, 09 Jan 2026 13:07:08 +0000

For humans org-mode is good at this

New comment by jodleif in "Over fifty new hallucinations in ICLR 2026 submissions"

jodleif — Sun, 07 Dec 2025 16:59:35 +0000

I find this to be a bit “easy”. There is such a thing as bad tools. If it is difficult to determine if the tool is good or bad i’d say some of the blame has to be put on the tool.

New comment by jodleif in "Coffee linked to slower biological ageing among those with severe mental illness"

jodleif — Sun, 07 Dec 2025 11:02:12 +0000

I assume these needs to be indications on overall diet.

Edit: i.e a bacon eater consumes a higher than average caloric intake, hazelnut eaters have more greens/vegetables in their diet possibly

New comment by jodleif in "The differences between an IndyCar and a F1 car"

jodleif — Sat, 06 Dec 2025 20:23:03 +0000

The huge difference with f1 is that each team more or less make their own _bespoke_ car (except engine). With indycar is much more of a spec series where Dallara(?) makes the chassies for each team.

New comment by jodleif in "DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]"

jodleif — Mon, 01 Dec 2025 22:11:40 +0000

A common pattern is to offload (most of) the expert layers to the CPU. This combination is still quite fast even with slow system ram, though obviously inferior to a pure VRAM loading

New comment by jodleif in "DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]"

jodleif — Mon, 01 Dec 2025 18:53:01 +0000

1. Have you seen the Qwen offerings? They have great multi-modality, some even SOTA.

New comment by jodleif in "DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]"

jodleif — Mon, 01 Dec 2025 18:51:48 +0000

Assuming your hardware premise is right (and lets be honest, nobody really wants to send their data to chinese providers) You can use a provider like Cerebras, Groq?

New comment by jodleif in "DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]"

jodleif — Mon, 01 Dec 2025 18:47:45 +0000

> Valuation is not based on what they have done but what they might do

Exactly what I’m thinking. Chinese models catching rapidly. Soon to be on-par with the big dogs.

New comment by jodleif in "DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]"

jodleif — Mon, 01 Dec 2025 15:56:09 +0000

I genuinely do not understand the evaluations of the US AI industry. The chinese models are so close and far cheaper