biotech offline. new age has begun. revolution is immanent.
biotech offline isn't about abandoning biology. it's about replacing growth through accident with growth through design. systems that remember themselves. infrastructure that can rebuild itself.
i’ve been thinking lately that there are moments where you don’t realize you’re crossing a threshold until you’re already standing on the other side of it. tonight felt like one of those nights.
i didn’t build anything flashy. no giant website appeared. no new application launched. there wasn’t a screenshot-worthy finish line. instead, i built infrastructure. the boring kind. the kind that quietly changes every project that comes after it.
i finally created homelab-dotfiles. it feels strange saying that because i’ve wanted something like this for months. one repository. one place where my environment lives. not just configuration files, but the philosophy of how i work. every machine, every container, every future server should feel familiar the second i log into it. arch. debian. ubuntu. proxmox. eventually it won’t matter. they’ll all speak the same language.
i spent most of the night building the foundation instead of the house.
forgejo now mirrors everything to github automatically. there’s something comforting about that. my own infrastructure remains home, but every commit quietly leaves footprints somewhere else too. redundancy has always felt less like paranoia and more like respect for future me.
ssh became its own little project. i generated a dedicated automation key instead of continuing to abuse my personal one. figured out how to stage everything through proxmox, documented the process, distributed it across containers. it’s funny how often the hard part isn’t writing software. it’s deciding how software should arrive in the first place.
then came the installer.
there’s something deeply satisfying about watching a fresh debian container slowly stop looking like debian and start looking like one of my machines.
os detection.
base packages.
git.
zsh.
neovim.
starship.
eza.
compatibility fixes that quietly smooth over the differences between distributions.
i rebuilt the installer several times across fresh containers until i trusted it. ct101. then ct102. then ct103. every failure taught me something small. every rerun became a little quieter. eventually the output stopped being a list of problems and became confirmation that everything was already exactly where it belonged.
that’s a satisfying feeling.
somewhere in the middle of all of that i ate chipotle. i also realized i’d been vaping almost continuously while debugging shell scripts. every compile, every syntax error, every successful run was punctuated by another reach for the vape. i noticed it more than i judged it. just another thing i’m trying to become aware of.
there’s still work left.
fastfetch.
dotfiles.
kitty.
btop.
my zsh configuration.
neovim.
eventually i’ll reach the point where i can create a brand new container, run a single installer, and watch it transform into something unmistakably mine.
that’s the real goal.
i called this entry biotech offline. new age has begun. because that’s what tonight felt like.
for years i’ve built things one issue at a time, solving the same problems over and over again. manually. almost organically. like every idea was growing independently.
tonight felt different. tonight felt like unplugging from that way of thinking. less hand-crafted biology.
more architecture.
more reproducibility.
more intention.
the machine that builds machines.
a bootstrap that creates the environment needed to build the next bootstrap. it’s a small thing on paper. a git repository full of shell scripts. but i think it’s the beginning of something much larger.
the old lab is slowly disappearing. the new one isn’t being assembled anymore.
it’s learning how to assemble itself.
in that process, she is also learning how to assemble herself.