06
The AI policy
Agents draft. Humans rewrite. No AI-written code or content ships without a senior engineer rewriting at least 30% of the words. No AI-generated hero images, ever.
Placeholder method entry. Rewrite the body before launch.
Our AI policy has two rules. First, agents draft and humans rewrite. Anything an agent produces — code, copy, a blog post — gets at least 30% of its words rewritten by a senior engineer before it ships. Second, no AI-generated hero images, illustrations, or stock-style imagery on the site, in client deliverables, or in marketing.
We use Claude Code, Roslyn-aware MCP servers, and structured hook pipelines. Heavily. The output is excellent. It is also recognizable, and the discipline of human rewriting is what keeps it from looking like everyone else’s recognizable AI output.
Why
The market in 2026 is saturated with AI slop. The differentiator is no longer “we use AI” — every agency does. The differentiator is the editorial floor that an AI workflow operates against. Ours is 30% rewrite plus zero generated imagery. Yours, if you hire us, becomes the same.
How we do it
- Every PR with AI-generated code is reviewed by a senior engineer who is responsible for the rewrite, not the agent.
- Every blog post draft is timestamped before and after the human pass; we measure the rewrite percentage.
- Hero images are screenshots, hand-drawn diagrams, or photographs. Never generated.
- We are willing to lose deals over this. We have.