Back to Insights

Substack

March 19, 2025

4 min read

Setting Up a New M4 Max MacBook Pro for AI Development

Henry

via Substack

View original

Why I Chose the M4 Max MacBook Pro

I’ve been an Apple user since 2007 when I got my first MacBook Pro in college. That was the moment I transitioned away from Windows as my primary OS and fully embraced UNIX-based systems. Since then, I’ve always had a MacBook as my daily driver, though I’ve used a few Windows machines, including my current one with an RTX 3090. I even had a high-end $15,000 desktop at one point, but dealing with fried processors and constant replacements made me realize that having a MacBook is my baseline requirement for development. While I’ve worked in environments like Verizon where Windows-specific builds were necessary, my work has always centered around UNIX/Linux-based machines.

As an AI developer, my workflow involves handling large language models, running complex DevOps pipelines, and optimizing blockchain consensus mechanisms. The new M4 Max MacBook Pro provides a powerful combination of performance, efficiency, and portability, making it an excellent choice for a machine-learning-heavy development setup.

This post details the command-line tools and IDEs I’ve installed to streamline my workflow.

Core Applications Installed

Beyond the command-line, I installed essential applications for productivity, note-taking, and development workflow.

General System Tools

• Homebrew - The backbone of my package management.

• Raycast - A powerful macOS launcher that replaces Spotlight.

• Ghostty / Warp / Hyper Terminal - Terminal replacements for a smoother, GPU-accelerated experience.

• Orbstack - Lightweight Docker alternative with fantastic MacOS integration.

• DropOver - Drag-and-drop utility to manage files seamlessly.

• Homerow Mac - Keyboard-first productivity booster.

• Fantastical / Notion Calendar - Calendar and task management.

Browser & Communication

• Arc / Chrome - Daily browsing.

• Linear - Task and issue tracking.

• Slack / Telegram - Communication tools.

• ChatGPT / Claude / Perplexity Desktop - AI-powered research assistants.

• Alex Sidebar - Sidebar utility for easier window management.

AI & LLM Development Tools

• LM Studio / Ollama / GPT4All - Running local LLMs for offline AI development.

• Satyrn Notebook - Notebook-style AI experimentation.

• Limitless AI / Granola - AI-powered workflow automation.

Coding & IDEs

• Zed / Zed Preview - High-performance text editor.

• Cursor / Trae / Windsurf - AI-assisted coding.

• TablePlus - Database management tool.

• Rive Desktop - Motion graphics editing.

• Xcode / Testflight - Native macOS and iOS development.

• Ulysses Notes / Bear Notes / Obsidian / Notion / Tana - Knowledge management and research.

Command-Line Tooling

A good macOS developer setup isn't complete without powerful CLI utilities. Below are my essential command-line tools, categorized by their functionality.

System Utilities

• bat - Better cat alternative with syntax highlighting.

• fd - Faster alternative to find.

• fzf - Fuzzy finder to speed up searching through history and files.

• eza - Modern replacement for ls with better visuals.

• ripgrep - Lightning-fast searching.

• mcfly - Smart command history search.

• zoxide - Smarter cd for navigating directories efficiently.

• sevenzip - File compression tool.

• uutils-coreutils - Rust-based replacements for GNU core utilities.

Development & Programming

• LLVM & LLD - Compiler infrastructure and linker.

• Solidity - Compiler for writing Ethereum smart contracts.

• Zig - New-age low-level programming language.

• Z3 - Theorem solver used in verification and AI/ML tasks.

• Helix - Fast and modern text editor.

• jq - Command-line JSON processor.

• Typst - Modern alternative to LaTeX for document creation.

• Git / JJ - Version control tools.

• GH - GitHub CLI.

• jless - JSON pager.

• SD - Stream editor that simplifies text replacement.

• LSD - Improved ls command with color coding.

Security & Networking

• GnuPG / GPGME - Encryption tools for secure communication.

• libssh / libssh2 - SSH libraries.

• OpenSSL@3 - Cryptography and security toolkit.

• Nmap / Rustscan - Network scanning tools.

• Unbound - Local DNS resolver for privacy.

Multimedia & Graphics

• FFmpeg - Command-line tool for video processing.

• ImageMagick - Image manipulation.

• Tesseract - OCR (Optical Character Recognition).

• WebP / FLAC - Compression tools for image and audio.

• Cairo - 2D graphics rendering.

• Freetype - Font rendering.

Data Processing & Compression

• SQLite - Embedded database for local storage needs.

• Brotli / Zstd / LZ4 / XZ - Various compression algorithms.

• libarchive - Archiving library.

• libb2 - BLAKE2 hashing.

• PCRE2 - Regular expressions processing.

Rust-Based Utilities (Cargo)

• Bottom (btm) - Modern system monitor.

• Dutree - Disk usage visualization.

• LSD - Enhanced ls alternative.

• SD - Alternative to sed for fast text replacement.

CLI Tools

• Doggo - Modern alternative to ping.

• DUA-CLI - Disk usage analyzer.

• Dust - Better du for checking disk usage.

• Erdtree - Enhanced tree command for directory visualization.

• GPing - Graphical ping tool.

• Tig - Git TUI (text-based UI for Git).

• Zenith - System resource monitor.

• Yazi - TUI-based file manager.

Fonts & UI Enhancements

• Font Symbols Only Nerd Font - Better terminal font support.

NPM Packages

• Corepack - Manages package managers (Yarn, PNPM, etc.).

• NPM - Node package manager.

Final Thoughts

This setup optimizes efficiency, AI development, DevOps workflows, and cryptographic security. The goal was to build an environment that supports:

• Running local AI models (LM Studio, GPT4All, Ollama)

• Smart contract development (Solidity, Z3, Zig, Rust tools)

• Advanced networking & security (Nmap, Rustscan, GPG, Unbound)

• Terminal-based productivity enhancements (Zoxide, FZF, Ripgrep, LSD)

• Database & backend tooling (TablePlus, SQLite, Brotli, Zstd)

This is just my initial baseline setup, and I'll continue to update it as I refine my workflow and discover new tools. This will also be the first post in a series, where I'll dive deeper into some of these tools and their use cases. If there's a particular tool you're curious about, let me know, and I can write a more in-depth post about it!