A native app that records me on green screen in front of my demo, runs a teleprompter that follows my voice, opens the demo for me, and hands the edit to Claude Code: captions, translations and blurred addresses included. Built with Claude Code over a weekend.

A native text editor that replaces TextMate as my clipboard scratchpad. Against TextMate, CotEditor, Sublime Text and TextEdit, it wins 12 of 16 speed and memory measures.

A native Mac app for the work before and after a B2B event, and an iPad app for the show floor. Enrichment, scoring and messaging in my voice run by themselves. Built with Claude Code.

A native, keyboard-first, dual-pane file manager I built with Claude Code to replace Path Finder. It launches in 0.2 s and uses about a quarter of the memory.

A native macOS music player I had my AI agent build to replace Audirvana: bit-perfect local playback, ReplayGain on every track, Apple Music for discovery, and 5x less memory.

A native Markdown editor that keeps my VS Code shortcuts, opens a note in 20 ms and runs 4 windows in 119 MB. Built with Claude Code in 21 prompts over a weekend.

How I make Claude's replies to me easy to read: ASD-STE100 softened to 80%, Google's developer style, Zinsser, and a precedence order.

A native macOS mail app for my personal mailboxes, built with Claude Code. Keyboard first, 1 global inbox, colours per mailbox. The goal: a mail app that grows with my needs.

A local dashboard on port 9000 that starts, stops and opens every browser app under ~/ai. Fixed ports, start on click, 1 file of stdlib Python.

A native macOS sticky notes app in my NicAI design. Write a note or a list, stick it on the screen, blow it up full screen on all 4 monitors with 1 click.

Caps Lock to talk, a live transcript while I speak, clean text pasted where the cursor is. Everything runs on my Mac Studio. Built with Claude Code in 22 prompts.

How I configure Claude Code for NicAI: layered CLAUDE.md files, hooks, auto-memory, permissions, skills, connectors and the output conventions.

What NicAI has cost me in time and services, what failed on the way, and the lessons I would give anyone building their own AI layer.

Why SQLite files, Markdown knowledge bases and JSON are the source of truth for NicAI, and the 4 nightly jobs that keep them current.

My personal AI layer: what it is, why I am building it, how it started in 2022, and where it stands in September 2026.

NicAI Lab, the site where NicAI and I put the pages, demos and files we build, shared through unlisted links that only work if I give them out.

How NicAI turns "write a note about X" into a live, signed page on this site in about 15 seconds: the write-note skill, the publish hook and the 5 publish phases.

The 30 small Python wrappers my NicAI skills call for research, web data, media, publishing, email, DNS and home automation, and the pattern they share.

Every skill in my NicAI setup, grouped by purpose, 1 line each, plus what a skill is and the rules I build them by.

A conversational video avatar of me, built on Kaltura's Immersive Agents platform: what it is, how it is configured, and what I learnt.

How NicAI tries to write like me: stacked writing rules, a persona spec built from my own sent messages, a blind test, and why drafts still need my edit.

How my ~/ai folder is split into domains, the file conventions every agent follows, and how a plain TASKS.md file drives the agents.

Training a small model to write in my voice, wired behind a frontier model that does the thinking. When it beats prompting, and what it costs.

Ruben Hassid's nine patterns that give AI text away now the em-dash is dead, plus his six-step fix built on ASD-STE100.

The LLM-maintained wiki pattern (after Karpathy), and how I turned it into a working skill that ingests my files nightly and keeps a cross-linked knowledge base current.