Hacker News: oofbey

New comment by oofbey in "Rio de Janeiro's "homegrown" LLM appears to be a merge of an existing model"

oofbey — Sun, 14 Jun 2026 20:06:23 +0000

Correct. We used to think that because NN optimization is non-convex there are all these local minima. Now we know that once you get past the very early parts of training from random init, the loss surface is fairly smooth, and not really convex, but close enough in a bunch of ways - linear combinations of trained models are pretty much always valid combinations. You can think of fine tunings as deltas on the original model which can be summed together successfully. I think this paper first showed that to me: https://arxiv.org/pdf/1802.10026 which was 8 years ago now.

New comment by oofbey in "πFS"

oofbey — Wed, 10 Jun 2026 22:40:39 +0000

That’s a lot of negatives!

New comment by oofbey in "xAI is looking more like a datacentre REIT than a frontier lab"

oofbey — Tue, 09 Jun 2026 13:47:24 +0000

Nice breakdown. But data centers are increasingly responsible for their own electricity. As in Colossus.

New comment by oofbey in "Siri AI"

oofbey — Mon, 08 Jun 2026 22:39:35 +0000

It’s definitely for hardware reasons. They have been aggressively improving the vector math capabilities in their chips, but as anybody who has tried to run a local LLM will tell you, newer hardware works better and you’re always limited in what you can do.

New comment by oofbey in "DeepSeek V4 Pro beats GPT-5.5 Pro on precision"

oofbey — Mon, 08 Jun 2026 04:21:08 +0000

Omg that is brilliant. I am so using this.

New comment by oofbey in "Nvidia is proposing a beast of a CPU system for Windows PCs"

oofbey — Sun, 07 Jun 2026 05:31:40 +0000

I cannot fathom why he brings up CPU SIMD as a potential comparative weakness on the NVIDIA Spark when it has teraflops of CUDA sitting right there.

New comment by oofbey in "pg_durable: Microsoft open sources in-database durable execution"

oofbey — Sat, 06 Jun 2026 02:58:27 +0000

Understanding and deciding to rely upon are not the same. Complexity is one of the biggest challenges in real world software systems. Keeping a subsystem’s responsibilities simple can do a lot for a system’s reliability. Most teams have extremely mature systems for managing complexity in their code - tests, CICD etc. Sure you CAN build all those things for your database too, but it’s more work. Most teams I’ve worked on choose to minimize database migrations because it’s a lot of work to make that part of the system as robust against mistakes as code is. Choosing to ignore a feature in your database is often a very rational pragmatic choice aimed at keeping complexity under control.

New comment by oofbey in "pg_durable: Microsoft open sources in-database durable execution"

oofbey — Sat, 06 Jun 2026 02:47:46 +0000

You could. But then you’re also building from scratch HA failover, backups, replica management, monitoring, etc - cloud vendor managed RDBMS come with lots of niceties. All of which are possible to set yourself. But a hassle, and difficult to make bullet proof.

New comment by oofbey in "The Ü Programming Language"

oofbey — Thu, 04 Jun 2026 03:14:42 +0000

You underestimate the egos involved.

New comment by oofbey in "They’re made out of weights"

oofbey — Thu, 04 Jun 2026 02:51:07 +0000

I love this. For anybody not getting the joke, it’s riffing on the classic 1990s essay “They’re made out of meat.”

https://web.mit.edu/people/dpolicar/writing/prose/text/think...

New comment by oofbey in "A walking tour of surveillance infrastructure in Seattle"

oofbey — Tue, 02 Jun 2026 14:42:50 +0000

They clearly have an agenda, but also openly acknowledge that public surveillance is a two sided coin, balancing public safety and convenience with privacy. Some of the risks they identify are real, but others are unabashedly exaggerated.

New comment by oofbey in "A walking tour of surveillance infrastructure in Seattle"

oofbey — Tue, 02 Jun 2026 14:40:42 +0000

Correct. All major OSes stopped broadcasting the preferred SSID list by 2017, with Android and Linux being the last. Apple stopped in 2014. Windows by 2009.

New comment by oofbey in "Corporate America Is Starting to Ration AI as Cost Skyrockets"

oofbey — Sat, 30 May 2026 14:28:59 +0000

I’m sorry you are used to working with out of touch leadership. Not all companies are like that. Even big ones can have smart, empathetic leaders. Although very often money gets in the way of empathy.

New comment by oofbey in "Anthropic surpasses OpenAI to become most valuable AI startup"

oofbey — Sat, 30 May 2026 14:13:23 +0000

GPT 5.5 still invents facts rather than looking them up, and manages to come across both as condescending and sycophantic. It feels like talking to a used car salesman.

New comment by oofbey in "Disagreement among frontier LLMs on real-world fact-checks"

oofbey — Thu, 28 May 2026 15:26:14 +0000

In many cases “I don’t know” is the correct answer - for questions about events that happened after the training cut off, if it doesn’t have web search, that is undeniably the correct answer. You’re forcing it to guess unnaturally. That really feels like you’re trying to prove a point (that your service can’t be replaced by AI) instead of actually performing research into how AI can be helpfully applied to this topic.

New comment by oofbey in "CISA tries to contain data leak"

oofbey — Sat, 23 May 2026 02:03:17 +0000

Willful ignorance. "No indication" meaning they haven't seen any evidence anything was compromised. Could be because they've been working very hard not to look at any evidence or analysis of what happened. "I'm not aware of X" is very different from "X is not true".

New comment by oofbey in "Google's Antigravity bait and switch"

oofbey — Fri, 22 May 2026 11:37:31 +0000

Classic Google problem: nobody gets promoted for maintaining anything. Promotions require doing Something Big. Convince everybody that your new feature is so much better than the old experience that it’s worthy of nuking the old experience, and that’s evidence that your New Thing is more worthy of you getting promoted. That’s how shit like this happens at Google.

New comment by oofbey in "The IBM-ification of Google?"

oofbey — Fri, 22 May 2026 11:21:37 +0000

As a former Googler this doesn’t surprise me at all. Dealing with customers directly is literally seen as failure. Everything at Google is about scale, and automation. If you have to do something manually - anything - then you’re doing it wrong. It’s deep in the culture, or was when I was there. Maybe they’ve gotten better, but no signs this is true.

New comment by oofbey in "Railway Blocked by Google Cloud"

oofbey — Wed, 20 May 2026 03:02:50 +0000

Railway deserves a lot of blame here. Deleting backups along with the database is a lot like not having backups. Moronic design choice.

New comment by oofbey in "Incident Report: Railway Blocked by Google Cloud [resolved]"

oofbey — Wed, 20 May 2026 03:01:27 +0000

After reading about how their delete database API also deletes all the backups, I concluded they are not to be trusted.