The build log · newsletter
Steal my automations.
One agent I built, the workflow behind it and the tools that made it work. Copy, paste, ship.
10,000+ builders already subscribed. Free.
- /01
Real builds
The agents and Claude Code setups I actually run, and the logic behind them.
- /02
Real numbers
What it cost, what it saved, what broke. Not just the success stories.
- /03
No fluff
Short, practical and written by me. No recycled AI news.
- /31 oct 2026
AI Agent Security: Limit What the Agent Can Do, Not What It Reads
AI agent security in plain terms: why prompt injection can't be filtered away, how to scope access, which actions need approval and what to log.
- /30 oct 2026
Claude Code GitHub Actions: setup, workflows and safe limits
How to set up Claude Code GitHub Actions, the three workflows worth starting with, what a run costs, and the limits that keep it safe in a real repo.
- /29 oct 2026
Agentic Workflows: 5 Patterns and How to Pick the Right One
Agentic workflows put an AI model inside a set of steps you define. The five patterns, when each one fits, and how to design your first one safely.
- /28 oct 2026
AI Operating Model: Who Owns, Runs and Fixes AI in Your Company
An AI operating model sets who owns each AI process, who builds and approves it, and how it is run and fixed. Six parts, three setups, how to start.