Posts
Posts
AI-built tools look done. They aren't.
AI assembles your client's tool in minutes and it looks ready. What it hides is where it breaks: the business logic, the security, and your brand on the front.
read →Pricing a client's tool: margin isn't the cheap dev
Your margin on reselling a tool doesn't come from the cheapest dev. It comes from a predictable delivery that doesn't come back broken on you.
read →I inherited a vibe-coded MVP to scale: the honest diagnosis
Inherited a stuck vibe-coded MVP to scale? Rewriting from scratch is the costliest mistake. The honest diagnosis: what to save, what to rewrite, how to stabilize.
read →A client wants a custom tool and your agency can't build it
The request landed, it's new revenue, and your team froze. Why a dev shop is expensive, an in-house hire is worse, and what the third way is.
read →Your MVP doesn't turn to junk for being fast, but for cutting the wrong thing.
The MVP that turns to junk wasn't built too fast. It merged everything to save time and kept the feature nobody uses. What to cut and what to keep.
read →Your codebase is the new prompt: the MVP that scales (or turns to junk)
In an AI-built MVP, what makes it an MVP that scales isn't the stack. It's whether the coding agent still navigates your repo six months later.
read →Spec-driven development: getting unstuck from vibe coding
Why vibe coding stalls the moment you try to scale, and how spec-driven development delivers an order of magnitude less rework with AI agents.
read →The spec was the easy part. SDD's bottleneck is execution
Spec-Driven Development externalizes intent into markdown, but pushes the pain into execution. Orchestrate with state outside the model's window.
read →Maestro + Claude Code: your app tested in the simulator like Playwright tests the web
You can have Claude navigate and test your app in the simulator like Playwright does on the web. With Maestro: one YAML, iOS and Android, zero instrumentation.
read →Code review became the bottleneck. CodeRabbit won't save you alone
A team went from 2 weeks of PR backlog to zero human review on staging. How CodeRabbit works, where it breaks, and why CLAUDE.md is the single source.
read →MCP Playwright: real-quality local validation before the PR
Before asking for a human review, Claude with MCP Playwright has already navigated your app, taken screenshots, and flagged regressions. Local, on your dev, in seconds.
read →Claude Code superpowers: the plugin that changes how a team ships
Claude Code superpowers isn't just another AI plugin. It's how you encode team knowledge into a place the machine reads and executes every time.
read →ADRs in Notion, without the bureaucracy
ADR isn't corporate ceremony. It's the cheapest way to stop repeating the same mistakes, and to make Claude Code consult your decisions before coding.
read →