Michael Wills michaelwills

The Software Factory Playbook — Dex Horthy's 4-gate workflow as a skill

A Claude Code Agent Skill built from Dex Horthy's (HumanLayer) playbook on David Ondrej's podcast.

"Once the model has written thousands of lines of code, it is harder to change. The sessions that generate design docs are context-light — you get the most model intelligence when you do the hard thinking early."

By default, agents build horizontally: all the backend, then all the frontend, then a 2,000-line diff lands in your lap and reviewing it is your problem. This skill flips that. Every decision that matters gets made before the code exists — where changing your mind costs a sentence, not a rewrite.

What it does

Setup

On every machine in the cluster install openmpi and mlx-lm:

conda install conda-forge::openmpi
pip install -U mlx-lm

Next download the pipeline parallel run script. Download it to the same path on every machine:

SynthLang: A Hyper-Efficient Prompt Language for AI

SynthLang is a hyper-efficient prompt language designed to optimize interactions with Large Language Models (LLMs) like GPT-4o by leveraging logographical scripts and symbolic constructs. By compressing complex instructions into fewer tokens (reducing token usage by 40–70%), SynthLang significantly lowers inference latency, making it ideal for latency-sensitive applications such as high-frequency trading, real-time analytics, and compliance checks.

Additionally, SynthLang mitigates English-centric biases in multilingual models, enhancing information density and ensuring more equitable performance across diverse languages. Its scalable design maintains or improves task performance in translation, summarization, and question-answering, fostering faster, fairer, and more efficient AI-driven solutions.

Large Language Models (LLMs) such as GPT-4o and Llama-2 exhibit English-dominant biases in intermediate embeddings, leading to inefficient and oft

Note

I’ve written a newer optimized 2026 version of this guide with a faster setup using Zinit + Starship and improved plugin loading.

This version avoids slow shell startups and works better with modern Zsh setups.

Updated guide: https://gist.github.com/n1snt/2cccc8aa5f7b645a7628d3512c70deb6

	import asyncio
	import time
	import logging
	from threading import Thread
	from nats.aio.client import Client as NATS
	from nats.aio.errors import ErrConnectionClosed, ErrTimeout

	class Component:
	component = None

	{
	"name": "gsap-to-video",
	"version": "1.0.0",
	"main": "index.js",
	"license": "MIT",
	"dependencies": {
	"fs-extra": "^7.0.0",
	"puppeteer": "^1.7.0"
	}
	}

	"""
	This gist shows how to run asyncio loop in a separate thread.
	It could be useful if you want to mix sync and async code together.
	Python 3.7+
	"""
	import asyncio
	from datetime import datetime
	from threading import Thread
	from typing import Tuple, List, Iterable

	{
	"name": "gsap-to-video",
	"version": "1.0.0",
	"main": "index.js",
	"license": "MIT",
	"dependencies": {
	"fs-extra": "^7.0.0",
	"puppeteer": "^1.7.0"
	}
	}

	<!doctype html>
	<html class="no-js" lang="en">
	<head>
	<meta charset="utf-8" />
	<meta name="viewport" content="width=device-width, initial-scale=1.0" />
	<title>Foundation \| Welcome</title>
	<link rel="stylesheet" href="css/foundation.css" />
	</head>
	<body>
	<div class="row">