10 things learned from AI code assistants

Benj Edwards tried numerous AI code assistants and came away with 10 lessons learned.

Claude Code, Codex, and Google’s Gemini CLI, can seemingly perform software miracles on a small scale. They can spit out flashy prototypes of simple applications, user interfaces, and even games, but only as long as they borrow patterns from their training data. Much like a 3D printer, doing production-level work takes far more effort. Creating durable production code, managing a complex project, or crafting something truly novel still requires experience, patience, and skill beyond what today’s AI agents can provide on their own.