idk exposing your home network to the world and trusting AI will produce secure code is not a risk I want to take

I just recently wrote my own agent that can gdb, objdump, nasm, cc, make, and more.

Agents are powerful. Even more so with skills and command line tools they can call to do things. You can even write custom tools (like I did) for them to use that allows for things like live debugging.

The tailscale piece to this setup is key.

My idea of fun is deeply tied to understanding how things work—learning them, then applying that knowledge in my own way, as simply as possible. That process gives me a sense of ownership and control, which is not something I get from an approach where AI does things for me that I do not understand.

I also liked using AI agents to do sysadmin stuff, especially with Nix OS. On top of Nix being great, the configuration of a system being files gives the agent good context on the current state the system is. Then when it does make changes, its great to be able to review its work via diffs.

I have found that storage is up in price more than 60% from last year.

I am writing a personal application to simplify home server administration if anybody is interested: https://github.com/prettydiff/aphorio

I have a hard time reading things like “The last one is the real unlock.” or “That alone justified the box.” without immediately thinking of an AI trying to explain something. Not to say this was written with one, but the frequency with which I see phrasing like this nowadays is skyrocketing...

Self hosting post. Tailscale.

Its comedic at this point.

My non-technical friend, never learned coding, doesn't know Linux, zero sysadmi experience does this and he can do anything and doesn't even know what Clause is doing. He learned some concepts recently like Docker, SSH, but that's basically it.

I think this is a good idea so long as you ensure you've got a good backup going or don't put anything super critical on there. I think it's seriously outside odds that Claude `rm -rf /`s your server, but definitely not 0%.

My workflow is a bit different in the sense I open my claude session in my laptop, at the directory of my ansible homelab code, and I also give Claude access to ssh to my homelab. But at the end it's almost the same, great tool.