13 posts
A comprehensive guide to maximizing LLM inference performance on Apple Silicon — MLX vs llama.cpp benchmarks, quantization formats, RAM requirements, MoE models, speculative decoding, KV cache optimization, and the best models for every Mac configuration.
A complete guide to building Andrej Karpathy's LLM Wiki — the AI-maintained knowledge base pattern that replaces RAG with structured markdown. Covers the three-layer architecture, setup with Claude Code + Obsidian, ingest/query/lint workflows, and community implementations.
A practical guide to decoding LLM model names — what B, IT, GGUF, Q4_K_M, MoE, and every other suffix means. Includes comparison tables, links to the best research resources, and real examples from Gemma 4, Qwen 3.5, and Llama 4.
A comprehensive comparison of 30+ tools for tracking AI token usage across Claude Code, Codex CLI, Cursor, Gemini CLI, and more — from lightweight CLI utilities to enterprise observability platforms.
A comprehensive guide to running LLMs locally — comparing 10 inference tools, quantization formats, hardware at every budget, and the builders empowering developers with open-weight models.
Master Mermaid diagram syntax for flowcharts, sequence diagrams, and more. Copy-paste examples, theming, and AI generation tips in one guide.
A comprehensive guide to 5 strategies for integrating Obsidian with Claude Code — vault organization, symlinks, MCP bridges, recommended plugins, and community best practices.
Part 2 of the CLI tools series — 10 more terminal tools to supercharge your AI coding workflow, plus resources for discovering new packages across Homebrew, NPM, and Rust crates.
A complete guide to installing and using Pixelmuse CLI for AI image generation from the terminal — from setup and authentication to prompt templates, MCP server integration, and Claude Code skills.
A complete guide to Yazi — the async, Rust-powered terminal file manager with image previews, Lua plugins, and vim-style keybindings. Covers installation, commands, plugins, and practical workflows for development, Linux administration, and AI-assisted coding.
A complete guide to symbolic links (symlinks) — what they are, how they work, the ln -s command, common gotchas like the trailing slash trap, and real-world use cases for dotfiles, Obsidian vaults, and dev workflows on macOS and Linux.
A complete guide to installing and using 10 terminal tools that pair perfectly with AI coding agents like Claude Code — from LazyGit for reviewing AI-generated diffs to Zoxide for lightning-fast directory navigation.
How real teams use Claude Code at scale: case studies from incident.io, Nx, Anthropic, Every, Y Combinator startups, Treasure Data, and best practices for parallel agents, git worktrees, and verification loops.
