Sounds interesting. A quicker, lighter java is better.

Needs a vibecode disclaimer

I’m taking the same approach for my main work laptop. As long as X11 is working, I probably won’t bother moving to wayland/sway.

Kiss, kvaesto (?), unlauncher.

I’ve been using unlauncher for about a year or so. It’s great.

Don’t waste time. Either neovim or emacs.

Youre doing something wrong but if plasma works for you, it works for you.

I cannot imagine using anything other than a tiling WM. The same way I cannot stand coding/writing without vim mode. Because I have become so efficient and good at them, things just get done way more quickly.

But if the WM doesn’t have that multiplier effect for you, just go tge easy path.

i3wm-ist here. Niri also didn’t provide anything to me i3 didn’t. It was actually more dificcult and less productive for my professional work with two monitors. But for laptop only usage it’s amazing.

i3 to sway is pretty smooth. You don’t have to change much and after a while you forget if youre using i3 or sway.

Movim seemed cool

Musicolet or vlc on my android. 100GB of local music on my SD card I have collected since I was a kid.

I don’t like the term “AI coding”, but recently I have been playing with opencode with it’s hosted models and ollama for light local models. It’s good as a rubber ducky, an assistant, or co-programmer.

What the actual fuck is this timeline that we are living in?

Qwen 3.5 24B is way too large for my specs. I’m barely running qwen2.5 7B

I started working toward self hosting LLM for my small company using ollama and opencode as agent But I realized a good model like GLM 5 requures 250GB of RAM and 24GB vram with a 4080?? I dont know, this is what the LLM told me itself.

I ended up using qwen-code2.7-7b-16k.

Currently the best thing I have is my laptop, 16GB ram, i7 9750H gtx1650

How do you guys selfhost? What models do you use that are actually good?

If you’re a GUI guy, catfish is good. If you’re a term guy, fzf, rg, fd are very fast and powerful. I search for files in / (meaning EVERYTHING) and tge results are there in less than 10 seconds.

This blog post has some example usages, read the entries for fzf, rg and fd.

https://alavi.me/blog/cool-cli-tui/#fd