Hacker News: ex1fm3ta

New comment by ex1fm3ta in "Mistral now trains on user input by default, except on enterprise tier"

ex1fm3ta — Wed, 02 Sep 2026 14:05:42 +0000

I mean they all do it right ? Mistral is just the first one to publicly say it.

New comment by ex1fm3ta in "Coding expertise is going to collapse from AI reliance"

ex1fm3ta — Tue, 25 Aug 2026 18:49:59 +0000

hmm that's a bit low right now, since you need at least 48Gb of VRAM to be confortable. 32Gb might do it (on quantitized models + optimized) but it would be very slow.

New comment by ex1fm3ta in "New Mac mini, featuring M6 and M5 Pro"

ex1fm3ta — Tue, 25 Aug 2026 18:46:15 +0000

yep why not ? "Unified Memory" was always their favourite selling argument

New comment by ex1fm3ta in "New Mac mini, featuring M6 and M5 Pro"

ex1fm3ta — Tue, 25 Aug 2026 14:42:00 +0000

same spec, same chip. I tought Apple would at least bump the RAM

New comment by ex1fm3ta in "Coding expertise is going to collapse from AI reliance"

ex1fm3ta — Mon, 24 Aug 2026 17:21:13 +0000

Keep in mind also that LLMs are currenctly heavily subsidized. Once VCs money are gonna run out, you will see the real price, and your 200K per year dev is probably nothing.

New comment by ex1fm3ta in "OpenAI: GPT 5.6 Sol price reduction (until at least Nov 21)"

ex1fm3ta — Mon, 24 Aug 2026 15:39:27 +0000

The Chinese are coming after these greedy-ass frontier labs. Today Xiaomi unveiled it's own inference machine .... I bet it's gonna be cheaper than Nvidia DGX, shipped with open source models that anybody can have at home.

New comment by ex1fm3ta in "Degraded performance for multiple models"

ex1fm3ta — Tue, 18 Aug 2026 17:05:34 +0000

I developed a small plugin for claudeCode that allows you to directly see in the console whats the status of claude-code in general and the status for your current model check => https://github.com/moumine9/claude-status

New comment by ex1fm3ta in "Tl;dv: Over 180k meetings left wide open"

ex1fm3ta — Mon, 10 Aug 2026 17:11:22 +0000

at this point, it's a feature, not a bug.

New comment by ex1fm3ta in "Discovery Loop"

ex1fm3ta — Wed, 05 Aug 2026 17:47:10 +0000

Sometimes you got to find a way to buy the silence of your top employee, to prevent them from going to the competition. This "start-up" is shallow as hell

New comment by ex1fm3ta in "OpenAI and Anthropic unite against open-weight AI risks to their bottom line"

ex1fm3ta — Thu, 23 Jul 2026 15:52:53 +0000

On today's episode of "AI Frontier Labs running scams": Now that the narrative of "too dangerous to be released" is not longer working since Chinese gave it all away, they have to come up with another one. You had the first glimpse this week: "Chinese frontier labs distilled our models ..." => so it should be illegal to use them in America since it's similar to illegal copy.

New comment by ex1fm3ta in "OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack"

ex1fm3ta — Wed, 22 Jul 2026 12:46:15 +0000

On today`s episode of "Ai Frontier Labs ongoing scams"

New comment by ex1fm3ta in "The Git history command deserves more attention"

ex1fm3ta — Tue, 14 Jul 2026 03:48:16 +0000

I do not agree at all. When you have multiple repos accross different services, commits are the best way to follow up with differents changes. We adopted conventional commits guidelines a few months ago and everyone is happy. Even ClaudeCode is able to keep up and auto fix stuffs with the proper commit messages. The changelog is dynamically updated. Everything so smooth when those commits messages are perfectly synchronized.

New comment by ex1fm3ta in "Grok CLI uploaded the whole home directory to GCS"

ex1fm3ta — Mon, 13 Jul 2026 14:07:46 +0000

Alex Karp was right, AI Compagnies are stealing people code while making them pay for unproductive tokens

New comment by ex1fm3ta in "What xAI's Grok build CLI sends to xAI: A wire-level analysis"

ex1fm3ta — Mon, 13 Jul 2026 13:35:09 +0000

So Alex Karp was right, AI companies "are stealing customers' data while charging them for unproductive tokens."

New comment by ex1fm3ta in "Canada is looking to build up to 10 new nuclear reactors over the next 15 years"

ex1fm3ta — Mon, 22 Jun 2026 20:01:16 +0000

unfortunately yes. Too much bullshits jobs (to suck up funds mostly and critize every aspects of non existant projects) and not enough people to take risks and do the job.

New comment by ex1fm3ta in "SpaceX Set to Overtake Amazon in Value as It Soars for Third Day"

ex1fm3ta — Tue, 16 Jun 2026 17:38:20 +0000

Fake money in a fake economy to buy out real companies.

New comment by ex1fm3ta in "Coreutils for Windows"

ex1fm3ta — Tue, 02 Jun 2026 17:42:30 +0000

I use those commands also to filter output and fee ai agents with that. Tail and Head are my favourites to avoid wasting tokens. Wayy too many fancy build logs messages.

New comment by ex1fm3ta in "Coreutils for Windows"

ex1fm3ta — Tue, 02 Jun 2026 17:40:47 +0000

nope. With ClaudeCode you can create skills ( basically markdown instructions) to teach your agents what command to use. You can also update CLAUDE.md to inject custom instructions that are feeded anything ClaudeCode is started.

New comment by ex1fm3ta in "ICE has spent over $25M on iris scanners in no-bid contracts"

ex1fm3ta — Thu, 28 May 2026 18:49:00 +0000

they did the same thing in Afghanistan and some scanners were found by the talibans once american soldiers left. Somebody even bought these from ebay and retro-engineered the software to access the data.

Edit: Article from New York Times https://www.nytimes.com/2022/12/27/technology/for-sale-on-eb...

New comment by ex1fm3ta in "Canada to order military plane fleet from Sweden in shift from US suppliers"

ex1fm3ta — Thu, 28 May 2026 14:04:18 +0000

Only C-Series program was sold to Airbus, Bombardier still makes private jet.