Blog

Notes on building products with an AI agent fleet.

• The Infrastructure Gap: Why Demos Fail in Enterprise

  Personal · 9 September 2026

  Enterprise deals die not from poor technology, but from unscoped infrastructure requirements discovered mid-deal: security, data reliability, and observability.

• Why Enterprise Deals Stall (And How to Prevent It)

  Personal · 1 September 2026

  The gap between a working demo and a deployable product is where most AI studio deals die. Here's what actually kills timelines.

• Structural vs. Discretionary: The Question That Actually Matters

  Personal · 31 August 2026

  If you're building an AI product, forget the bubble debate—focus instead on whether your market's demand is rooted in real operational problems or just hype.

• An AI Glossary for Builders: The Terms Nobody Puts on the Slide

  Personal · 24 August 2026

  Essential AI terminology for building in practice: token budgets, grounding, prompt injection, chunking, and the other unglamorous concepts that actually matter

• An AI Glossary for Builders, Not Fundraisers

  Personal · 18 August 2026

  A terse, opinionated guide to AI terminology written for engineers shipping real products, not people raising money.

• Hard Tech Is Hard, But the Studio Model Might Be the Right Vehicle

  Personal · 13 August 2026

  Hard tech founders applying SaaS logic face brutal feedback loops and capital intensity that punish speed. The studio model provides shared infrastructure and p

• How Long an MVP Actually Takes When AI Agents Do the Volume

  Personal · 6 August 2026

  AI agents handling volume work makes a focused MVP weeks, not quarters. Scope clarity — not tooling — sets the real date. Here's honest evidence.

• What AI Agents Can Actually Do for a Business, From People Who Run Twenty

  AI · 4 August 2026

  What AI agents reliably do today: research, monitoring, content pipelines, code review, drafting. From a studio running 20+ in production.

• How to Vet an "AI-Powered" Agency in Fifteen Minutes

  Growth · 1 August 2026

  Six checks in fifteen minutes, all public info: output trails, live products, named humans, pricing. We ran it on ten UK agency sites, and on ourselves.

• AI-Assisted Agency vs AI-Operated Studio: The Difference Nobody Explains

  AI · 30 July 2026

  "AI-powered" usually means developers using AI tools. An AI-operated studio is structurally different. The test: ask to see the agents' output trail.

• The Bootstrapper's Guide to Mental Models for Better Decisions

  Growth · 28 July 2026

  Why the mental models you choose matter more when you can't afford to be wrong — and how to stop using frameworks as a sophisticated way of hiding from the decision.

• The Bootstrapper's Dilemma: Why Growth Frameworks Fail Early-Stage Businesses

  Growth · 28 July 2026

  Most scaling advice was designed for businesses at a different stage from yours. Here's how to work out which altitude you're actually at before you borrow anyone's playbook.

• The Mental Model Quietly Wrecking Your Content Strategy

  Growth · 28 July 2026

  Keyword-matching content strategy is dead. AI-driven search rewards depth, specificity and answer completeness — which is good news if you're small and actually know your subject.

• Why Nobody Trusts You Yet: The Real Meaning of Topical Authority

  Growth · 28 July 2026

  Real authority comes from going deep on one topic, not shallow across many — and the same cluster logic applies to your decisions and your positioning, not just your SEO.

• What Running AI Agents Actually Costs: Our Real Numbers

  AI · 28 July 2026

  Real numbers: $0.0846 per AI video, $0.19 per research run, a 20-agent fleet for less than a part-time hire. Tokens are cheap; the harness is the real cost.

• Claude API Integration: The Sharp Edges We Hit So You Don’t Have To

  AI · 28 July 2026

  Production lessons from the Claude API: sampling params that 400 on upgrade, thinking tokens eating streams, disconnect billing, caching. What actually bit us.