Viewing profile — oakpond

>To support further research, we open-source the KDA kernel and vLLM implementations, and release the pre-trained and instruction-tuned model checkpoints. This is just awesome.

Shouldn't the energy instead be focused on creating a standardized eIDAS driver API that OS vendors are required to implement?

I notice this kind of "Sent from iPhone"-type spam with other AI tools too. It's awful.

This is a fp16 model. That's 54G in weights. I can load it only with fp8 quantization enabled (>= 128k context). I run into this error during generation though: https://github.com/…

Running dual Pro B60 on Debian stable mostly for AI coding. I was initially confused what packages were needed (backports kernel + ubuntu kobuk team ppa worksforme). After getting …

Working for what? Can you show some complex systems that are built with it? Their site only mentions a kanban board app and a photo album. I can believe it works for that.

I build with LLMs all the time but I generally don't do vibe coding unless it's something small I don't really care about. When I look at SpecKit, I see a kind of vibe coding fanta…

In my experience, when you sell expensive complex systems, customers are very worried about any differences in system behavior as a result of software updates. When you implement a…

I think natural language leaves too much room for ambiguities. If you treat it as code I expect you will run into frequent bugs and unintended side effects of LLM-authored changes …

Sort of agreed. Natural language specs don't scale. They can't be used to accurately model and verify the behavior of complex systems. But they can be used as a guide to create for…

probably yes. considering that even some of their non-ocr models can recognize my shitty handwritten math

It makes sense to me as long as you're not vibe coding the PBTs.

You do. Even the latest models still frequently write really weird code. The problem is some developers now just submit code for review that they didn't bother to read. You can tel…

Hardly anything substantial about how well this works in practice. It's a hiring ad.

> There's a new programmable layer of abstraction to master (in addition to the usual layers below) involving agents, subagents, their prompts, contexts, memory, modes, permissions…

The first two problems can be solved in a straightforward way with more custom exception types. For the "bigger problem", catch(...) can be used to prevent your code from crashing.…

I recommend trying to implement your own Forth in assembly. It's a great educational experience. However, I would not choose to use it in production. It's not a readable syntax. I …

I'm on Sway and Debian. Debian has a separate package for KDE Connect: https://packages.debian.org/trixie/kdeconnect

I've been using KDE Connect without KDE for a while now. Works for me (tm)

What makes you think they will? What makes you think regulators don't also want this?

Goodbye Android.

This simply must happen. I don't think there is another sustainable solution to the network-effect problem.

You're not the only one. LLMs are hardly intelligent anyway.

I'm also not bullish on this. In the sense that I don't think LLMs are going to get 10x better, but they are useful for what they can do already. If I see what Copilot suggests mos…