My tech registry

This registry lists every single technology I have installed globally, so I can easily reproduce my environment and keep everything up to date.

Anything that's installed by brew doesn't need detailed instructions, just run brew update && brew upgrade && brew cleanup.

Browsers

/opt/homebrew/bin/brew

Installation: /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
Environment config: eval "$(/opt/homebrew/bin/brew shellenv)"

/opt/homebrew/bin/bash

/opt/homebrew/bin/git

Installation: .zip install
Commands: ~/.local/bin/code (custom-created symlink to /Applications/Visual Studio Code.app/Contents/Resources/app/bin/code)

GPG /usr/local/bin/gpg (symlink to /usr/local/MacGPG2/bin/gpg)

Thanks to Apple, my C/C++ setup is very complex.

Dafny /opt/homebrew/bin/dafny

GHCup ~/.ghcup/bin/ghcup
- Installation: curl --proto '=https' --tlsv1.2 -sSf https://get-ghcup.haskell.org | sh
- Update: ghcup upgrade
GHC/Stack/HLS/... ~/.ghcup/bin/*
- Installation: ghcup install <tool> <version>, ghcup set <tool> <version>
- List installed tooling versions: ghcup list
- Note: Prefer stack over cabal.
Haskell executables: ~/.stack/bin/*
- Installation/Update: stack install <package>
- Configuration: ~/.stack/config.yaml
- Alex alex
- Happy happy
- HLint hlint
- Ormolu ormolu

NVM:
- Installation: curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash
- Environment config
~/.nvm/versions/node/v<version>/bin/node
- Installation: nvm install <version>; nvm use <version>
/opt/homebrew/bin/deno
- Installation: curl -fsSL https://deno.land/install.sh | sh
- Environment config: . "$HOME/.deno/env"
- Update: deno upgrade
~/.bun/bin/bun
- Installation: curl -fsSL https://bun.sh/install | bash
- Update: bun upgrade
- Note: Prefer bun over npm/yarn/pnpm where possible. Prefer pnpm over npm/yarn for Node projects.
npm executables: ~/.bun/bin/*
- List installed: bun pm ls -g
- eshost-cli eshost
- esvu esvu
- Mermaid CLI mmdc
- pnpm
- prettier
- Pyright pyright
- serve serve
- yarn

juliaup ~/.juliaup/bin/juliaup
- Installation: curl -fsSL https://install.julialang.org | sh
- Update: juliaup self update
Julia ~/.juliaup/bin/julia
- Installation: juliaup add <version>; juliaup default <version>

TeX Live /usr/local/texlive/2025/bin/universal-darwin/*
- Installation: follow instructions
- Update: reinstall

elan ~/.elan/bin/elan
- Installation: curl https://elan.lean-lang.org/elan-init.sh -sSf | sh
- Update: elan self update
Lean ~/.elan/bin/lean*
- Installation: elan install <version>; elan default <version>

Installation: .dmg install
Commands: /Applications/Racket v<version>/bin/racket
Update: Download latest version and install over existing installation. Update PATH as well.

OPAM: /opt/homebrew/bin/opam
OCaml: ~/.opam/<version>/bin/ocaml*
- Installation: opam switch create <version>
- Environment config: eval $(opam env)
- Update: opam update && opam upgrade
- List installed versions: opam list
OPAM executables: ~/.opam/<version>/bin/*
- Rocq rocq
- vsrocq-language-server
- OCamlFormat ocamlformat

perlbrew ~/perl5/perlbrew/bin/perlbrew
- Installation: curl -L https://install.perlbrew.pl | bash
- Update: perlbrew self-upgrade
Perl ~/perl5/perlbrew/perls/<version>/bin/*
- Installation: perlbrew install <version>; perlbrew switch <version>
- Environment config: . "$HOME/perl5/perlbrew/etc/bashrc"

/opt/homebrew/bin/uv
Python ~/.venv/bin/python, ~/.local/bin/python<version>
- Installation: First run uv python install <version>; then use uv venv .venv at user root
- Update: uv python update <version>; then recreate venv with uv venv .venv --force
- Note: The venv uses the Python version used to create it. You can also create other environments with other Python versions, probably in project directories, by installing additional Python versions.
Python executables: ~/.venv/bin/*
- Installation: uv pip install <package>
- black
- jupyterlab
- You are free to install any package used by casual scripts, but the following should be the baseline installations:
- matplotlib (library)
- NetworkX networkx (library)
- numpy (library)
- pandas (library)
- scikit-learn (library)
- scipy (library)
- Seaborn seaborn (library)
- statsmodels statsmodels (library)
- tqdm tqdm (library)
/opt/homebrew/bin/conda
- Note: I don't use Anaconda. But there's one particular package, Montreal Forced Aligner, that to this day doesn't have a functional PyPI build. I installed miniconda for it, and a little shell setup:
```
conda_init() {
if [[ $(which python) == *".venv"* ]]; then
    deactivate
fi

# >>> conda initialize >>>
# ...
# <<< conda initialize <<<

}
```
bash
Now I can run conda_init in any shell to set up conda for that session, and then run MFA from there.

Air: /opt/homebrew/bin/air
R
- Installation: .pkg install to /Applications/R.app, /Library/Frameworks/R.framework/Resources/bin
- Commands: /usr/local/bin/R, /usr/local/bin/Rscript (symlinks to /Library/Frameworks/R.framework/Resources/bin)
- Update: Download latest version and install over existing installation.

rbenv /opt/homebrew/bin/rbenv
Ruby ~/.rbenv/shims/ruby
- Installation: rbenv install <version>; rbenv global <version>
- Environment config: eval "$(rbenv init - bash)"
Ruby executables: ~/.rbenv/shims/*
- Installation: gem install <package>
- Update: gem update <package>
- Linguist github-linguist

rustup ~/.cargo/bin/rustup
- Installation: curl --proto '=https' --tlsv1.2 https://sh.rustup.rs -sSf | sh
- Update: rustup self update
rustc/Cargo: ~/.cargo/bin/*
- Installation: rustup install <version>; rustup default <version>
- Update: rustup update

Wolfram Engine /Applications/Wolfram Engine.app
- Installation: .dmg install
WolframScript: /Applications/WolframScript.app
- Installation: .dmg install
- Commands: /usr/local/bin/wolframscript (symlink to /Applications/WolframScript.app/Contents/MacOS/wolframscript)

Other libraries: