dmesg --follow
[ 68673720.000 ] posts.x: @davidsenra you will love this!  |   [ 68673600.000 ] posts.x: https://t.co/vhJifvV8y9  |   [ 68673600.000 ] posts.x: This Jay-Z documentary, In 8, is exceptional! Nearly 6 hours of intimate conversation between @sc and the brilliant @RickRubin. It’s rare to get such…  |   [ 68662260.000 ] posts.x: Full talk on turning re-prompted fixes into linters, including the OpenAPI drift fix and the skill-writing whack-a-mole:  |   [ 68662260.000 ] posts.x: Re-prompting an agent to fix the same sloppy pattern over and over isn't a workflow, it's a sign the rule should live in code, not in your head…  |   [ 68650020.000 ] posts.x: ↻ @willcb: me: kinda fucked up how many people are offloading their whole lives to muse or instinct or grokbot or whatever claude: yeah  |   [ 68650020.000 ] posts.x: ↻ @iamdevloper: signs you've become the senior dev: you're the only one who understands the deploy script you're the only one who remembe…  |   [ 68649720.000 ] posts.x: Gooooooooood morning! Brazil Oberon Cerrado. Roasted 110 seconds past first crack. Milk chocolate and almond.  |   [ 68647800.000 ] posts.x: @Saanvi_dhillon “Git expert” is a weird flex, though, isn’t it?  |   [ 68610720.000 ] posts.x: @sujee_dev @dkundel @aiDotEngineer Thanks mate!  |   [ 68600580.000 ] posts.x: Full talk on the Codex app server protocol and what you can build on top of the Codex harness:  |   [ 68600580.000 ] posts.x: A lot of engineers assume building their own agent interface means writing their own agent harness from scratch. @dkundel, Developer Experience Lead…  |   [ 68564580.000 ] posts.x: Full talk on why Reducto deleted their all-in-one MCP server and rebuilt it around fewer, state-aware tools and visible agent uncertainty:  |   [ 68564580.000 ] posts.x: A single massive MCP file worked fine as a curated demo and broke the moment Reducto's own team got hold of it. @abhiarya, who works on product…  |  
corey@gallon.me:~/til$

A nice workflow for creating Python projects with AI

FIGURE 1 ⋅ A nice workflow for creating Python projects with AI

First, create a new project using poetry

poetry new asciinema-editor

This gives you a nice, clean new project structure.

.
├── README.md
├── asciinema_editor
│   ├── __init__.py
├── poetry.lock
├── pyproject.toml
└── tests
    └── __init__.py

Now, let the model know that you have this project structure and ask it to write you code. Claude does a nice job of providing the relevant bits to make your project work nicely.

  • e.g. __main__.py (see TIL on this)
  • Exposes the relevant main function in __init__.py making it easier to import and use the package in other scripts if needed.
corey@gallon.me:~$ tail -f /writing Attach to the stream. An email when I have something worth sending. Replies encouraged!
corey@gallon.me:~$ ls -lt /til ↑2024-11-14 Wrap with HTML tags in VSCode
▸2024-10-15 A nice workflow for creating Python projects with AI ⋅ you are here
↓2024-10-15 __main__.py file in a project