Developer Tooling
Getting Started with AI-Assisted Coding Tools
Three guides that cover the full spectrum of AI-assisted development, from first steps with Copilot and Claude Code to managing the psychological load of constant AI integration.
How to Become an Expert in AI-Assisted Coding — A Handbook
TLDR: A comprehensive handbook covering Copilot, Claude Code, and Gemini CLI. The core finding is that experienced developers see 3-4x productivity gains, while juniors benefit less because they lack the domain knowledge to verify AI output.
Key Insight: The productivity multiplier scales with experience — senior developers get the most gains because they can evaluate and steer AI output effectively.
How to Not Be Overwhelmed by AI — A Developer’s Guide
TLDR: A practical guide to engineering mental models for working alongside AI. The central concept is the “verification gap” — the distance between what AI generates and what a developer can confidently validate.
Key Insight: The verification gap is the most important concept for responsible AI development, and closing it requires deliberate skill-building, not just better tools.
Practical Use of AI Coding Tools for the Responsible Developer
TLDR: Two years of hands-on experience with Copilot, Cursor, and Claude at a design studio. The article documents real wins in test generation, shader creation, and legacy code navigation, alongside honest assessments of where the tools fall short.
Key Insight: AI shines brightest on tasks developers hate — tests, legacy code, and boilerplate — which makes adoption organic rather than forced.
What does this mean for developers?
AI-assisted coding rewards those who already understand their domain. The consistent finding across these guides is that AI tools amplify existing skill rather than replacing it. Developers who invest in verification practices and deliberate tool selection will compound their productivity gains over time.