AI

Bridging Strategy and Execution in the Age of AI Agents

AI changed what's fast. Nobody changed what's slow. Teams are merging twice as many pull requests while review times double and bug rates climb. The bottleneck moved — and most organisations are only now starting to notice.

Eight Editors, Zero Writers: Why I'm Building Scriptorium

I've finished books but never sent the fiction to agents — never felt it was ready. Twenty years later, I'm building the tool I wish I'd had: not an AI that writes for you, but one that reads like a professional editor.

Context Engineering: The Skill That Replaced Prompt Engineering

Everyone's talking about context engineering as if it's a prompting technique. It's not. It's the skill that determines whether your AI investment produces strategy or noise — and the part that matters most has nothing to do with code.

Diff: Code Review Context Your LLM Actually Needs

Pasting a raw git diff into your LLM is asking for a review with half the files missing. cxpak diff adds callers, callees, and type signatures — within a token budget.

Human-in-the-Loop: Why Full Automation Is the Wrong Goal

The best AI systems aren't autonomous. They're collaborative. Full automation is a seductive goal that breaks on contact with reality — the winning pattern is bounded autonomy.

Trace: Finding the Code That Matters Before Your LLM Sees It

Most LLM context is wasted on code that doesn't matter. cxpak's trace command walks your dependency graph and packs only the relevant paths — so your model debugs with a map, not a haystack.

Spending CPU Cycles So You Don't Spend Tokens: Building cxpak

Why I built a Rust CLI that packages codebases into token-budgeted briefings for LLMs — and what I learned about the real cost of context.

Setting Up Multi-Agent Workflows with Claude Code

What I learned building a trading platform backoffice with six AI agents — before the playbook existed.