AI for the Rest of Us

    ___    ________________
   /   |  / ____/_  __/ __ \
  / /| | / /_    / / / /_/ /
 / ___ |/ __/   / / / _, _/
/_/  |_/_/     /_/ /_/ |_|

===============================
  A I   f o r   T h e   R e s t
===============================

One-command setup to supercharge your work with AI coding tools - no programming experience required.

Who Is This For?

You're not a software developer, but you have real problems to solve at work. You need to analyze messy data, automate repetitive tasks, or build tools that don't exist yet. This project sets you up with AI assistants that can do the technical heavy lifting while you focus on the problem.

What Can You Actually Do?

With AI coding tools, non-programmers can tackle tasks that used to require hiring a developer:

• Data Analysis - Query large datasets, find patterns, generate reports
• Data Cleaning - Transform messy CSVs, deduplicate records, standardize formats
• Browser Automation - Scrape websites, fill forms, automate repetitive web tasks
• File Processing - Batch rename, convert formats, extract information from PDFs
• Custom Tools - Build small apps tailored to your specific workflow
• API Integrations - Connect services together, pull data from various sources

Why "The Rest of Us"?

This phrase comes from Apple's 1984 Macintosh slogan: "The computer for the rest of us." It meant computers weren't just for programmers anymore. The same shift is happening now with AI coding tools - you don't need to be a developer to build software. You describe what you need, and AI writes the code.

Why CLI Tools Over Chat Interfaces?

AI coding assistants like Claude Code, Codex, and Gemini CLI run directly in your terminal. They're fundamentally different from chatting with AI in a browser:

CLI Tools	Chat Interfaces
Read and write files directly on your machine	Copy-paste code back and forth
Execute commands and see real output	Guess at what might work
Understand your entire project context	Limited to what you paste in
Process your actual data files	Work with toy examples
Run scripts and fix errors automatically	Manual iteration loop
Works with your local files	Everything lives in the browser

The CLI tools turn AI from a "coding assistant you talk to" into a "coding partner that works alongside you" - it can see your files, run code, and iterate until the job is done.

Quick Install

This repository provides two main components:

1. Environment Setup - One-command installation of development tools and AI assistants
2. Project Tools - CLI for scaffolding data science projects

Environment Setup

Windows

Open PowerShell and run:

irm https://raw.githubusercontent.com/rlancer/ai-for-the-rest/main/scripts/setup.ps1 | iex

macOS

Open Terminal and run:

curl -fsSL https://raw.githubusercontent.com/rlancer/ai-for-the-rest/main/scripts/setup.sh | bash

After setup completes:

1. Open a new terminal window (Windows Terminal on Windows, iTerm2/Terminal.app on macOS)
2. Run: aftr setup
   • This will prompt you to select which AI CLI tools to install
   • Optionally configure SSH keys

Or for non-interactive setup with defaults (Claude Code only, skip SSH):

aftr setup --non-interactive

Project Scaffolding with aftr

After environment setup, use the aftr CLI to create data science projects with uvx (no installation required):

# Interactive mode
uvx aftr

# Create a new project directly
uvx aftr init my-project
uvx aftr init my-project --path /custom/path

What aftr creates:

my-project/
├── data/              # Input data (gitignored)
├── notebooks/         # Jupyter notebooks with example.ipynb
├── outputs/           # Output files (gitignored)
├── src/my_project/    # Python module (hyphens → underscores)
├── .gitignore
├── .mise.toml         # Python 3.12, UV latest
├── pyproject.toml     # pandas, polars, jupyter, papermill
└── README.md

Custom templates: aftr supports custom project templates with different dependencies and files. See the template documentation for details on creating and managing templates.

Requirements

Windows

• Windows 10/11
• Classic PowerShell (not PowerShell Core)

macOS

• macOS 10.15 (Catalina) or later
• Homebrew (will be installed automatically if not present)

What Gets Installed

Environment Tools

Windows (via Scoop)

Package	Description
duckdb	In-process SQL OLAP database
Hack-NF	Hack Nerd Font
mise	Polyglot runtime manager
pwsh	PowerShell Core
starship	Cross-shell prompt
touch	Create files
uv	Fast Python package manager
vscode	Visual Studio Code
which	Locate commands
windows-terminal	Modern terminal

macOS (via Homebrew)

Package	Description
bun	Fast JavaScript runtime
duckdb	In-process SQL OLAP database
font-hack-nerd-font	Hack Nerd Font
iterm2	Terminal emulator
mise	Polyglot runtime manager
starship	Cross-shell prompt
uv	Fast Python package manager
visual-studio-code	Visual Studio Code

AI CLI Tools

Selected interactively during aftr setup:

Tool	Command	Installation	Platforms
Claude Code	claude	Native installer	Windows, macOS
Codex	codex	Bun global	macOS only
Gemini CLI	gemini	Bun global	macOS only

Note: Claude Code (recommended) uses Anthropic's official native installer and works on both platforms. Codex and Gemini CLI require Bun, which is only installed on macOS.

Project Tools

aftr CLI (via uvx)

The aftr project scaffolding tool runs directly with uvx - no installation needed:

Command	Description
uvx aftr	Interactive project creation menu
uvx aftr init <name>	Create a new data science project
uvx aftr init <name> -p <path>	Create project at custom path

Included in scaffolded projects:

• pandas & polars for data analysis
• jupyter & papermill for notebooks
• pytest & ruff for testing and linting
• mise for tool version management
• UV for fast Python package management

Post-Setup

Environment Configuration

The setup scripts automatically:

• Configure shell profiles for mise and starship
• Install base tools (Python, UV, aftr)

You then complete the configuration by running aftr setup to select AI tools.

Windows

1. Open a new Windows Terminal window
2. Run aftr setup to select AI CLI tools (Claude Code, Codex, and/or Gemini)
3. (Optional) Configure Starship by creating ~/.config/starship.toml
4. Set your API keys:

   $env:ANTHROPIC_API_KEY = "your-key"
   $env:OPENAI_API_KEY = "your-key"      # If you selected Codex
   $env:GEMINI_API_KEY = "your-key"      # If you selected Gemini

5. Run claude, codex, or gemini (depending on what you installed) to start coding with AI

macOS

1. Open a new terminal window (iTerm2, Terminal.app, etc.)
2. Run aftr setup to select AI CLI tools (Claude Code, Codex, and/or Gemini)
3. (Optional) Configure Starship by creating ~/.config/starship.toml
4. Set your API keys:

   export ANTHROPIC_API_KEY="your-key"
   export OPENAI_API_KEY="your-key"      # If you selected Codex
   export GEMINI_API_KEY="your-key"      # If you selected Gemini

   Or add them to your ~/.zshrc for persistence.
5. Run claude, codex, or gemini (depending on what you installed) to start coding with AI