A Fractional CTO's Claude Code Playbook: How I Ship Client Work 3x Faster
The exact Claude Code workflow I use across every fractional CTO engagement — from planning to PR — and the guardrails that keep it safe for production.
Read →13 posts
The exact Claude Code workflow I use across every fractional CTO engagement — from planning to PR — and the guardrails that keep it safe for production.
Read →AI coding tools multiply the output of experienced engineers and often slow down inexperienced ones. Here's why the gap exists and what teams can do about it.
Read →An honest, fractional CTO's review of the AI-native developer tools in 2026 — what each one is best at, what they cost in productivity versus oversight, and what I actually use on client work.
Read →Claude Code becomes dramatically more useful when you configure it properly for your team. Here's how I use subagents, skills, and hooks to turn it from a single-player tool into a team-level force multiplier.
Read →AI coding tools are only as good as the spec you give them. Here's the practice of writing structured specifications that make AI-generated code predictable, reviewable, and correct.
Read →AI-generated code introduces specific security failure modes that standard code review misses. Here's the security-focused review checklist I use on every engagement where AI tools are producing production code.
Read →AI-authored pull requests fail in different ways than human-authored ones. Here's the specific review heuristics I teach teams to catch the mistakes AI makes.
Read →The Model Context Protocol lets coding agents talk to your databases, APIs, and internal tools. Here's a developer-level walkthrough of building and deploying MCP servers for real engineering workflows.
Read →There are at least six distinct ways to use AI in a development workflow, from autocomplete to autonomous agent swarms. Here's the taxonomy, what each pattern is good for, and which ones are production-ready.
Read →The quality of an AI coding agent's output depends on what it reads before it writes. Here's how to engineer the context — repo maps, symbol graphs, file selection — so the agent starts with the right information.
Read →An honest operational guide to running coding agents on real codebases — the tasks they handle well, the traps they fall into, and the guardrails that keep them from causing damage.
Read →A practical, section-by-section guide to writing a CLAUDE.md file that turns a generic AI coding agent into an engineer who has read the onboarding doc.
Read →TDD gets a second life with AI coding tools. Write the tests first, hand them to the agent, and the agent writes the implementation. Here's how that workflow actually plays out.
Read →