Writing
Stories from the build, and essays on control.
Short, true stories from shipping ScotiTech's products, and longer pieces on private AI, app distribution and software organisations can own.
Story · Craft · 3 min read
We built a film engine to make a 75-second film
Vamiq's launch film has no video editor behind it. Every frame is a web page, rendered on a Mac with nothing to install.
Read the story →Story · Shipping · 3 min read
Three releases in one day, and a guard on every page
On 21 October we shipped three releases of the Vamiq website, at 09:48, 13:35 and 21:42. An automatic check made sure none of them cost us a single search result.
Read →Story · Naming · 2 min read
Two days from AppDeploy to Vamiq
On a Monday I decided to rename our product. By Tuesday night it was live under its new name, on a new domain, and nobody had anything to redo.
Read →Story · Copy · 2 min read
Confident, not confessional
Our website was so careful to be honest that it kept apologising. Here is how we rewrote it to sound sure of itself, without a single line that isn't true.
Read →Essays
On private AI, distribution and systems.
- 9 June 2026Private AI Workspaces for Regulated OrganisationsWhy regulated teams need private AI workspaces with governance, data sovereignty, auditability, and operational control instead of unmanaged AI chatbots.
- 16 February 2026Why Internal Tools Fail After ScaleWhy internal tools fail when they become operational infrastructure, and how to design them with ownership, auditability, reliability, and governance from the start.
- 16 February 2026Systems Thinking vs Application ThinkingWhy operational software succeeds when it is designed as a system, with dependencies, governance, reliability, auditability, and long-term ownership in mind.
- 8 February 2026App Distribution Without Full MDM: A Technical Deep-DiveHow internal app distribution can work without full device enrollment, with access control, auditability, and a privacy-respecting architecture for BYOD environments.
- 15 January 2026Building Privacy-First AI SystemsArchitecture patterns for privacy-first AI systems with data sovereignty, local inference, retrieval governance, human review, and audit-ready controls.