AIA critique of eight AI coding habits, from to-do lists to code-quality worries
A user says they spoke with more than 50 people about coding with AI and lists eight practices they consider mistakes—including stressing over code quality and using branches and worktrees.
TLDR
After what they describe as several months of one-on-one conversations with more than 50 people, a user outlines eight AI coding practices they see as mistakes: talking to 2026 models like they’re 2025 models; devising “funky metaphorical factories” to orchestrate agents; treating context as sacred; treating previous conversations as sacred; having a to-do list; using “funky tools”; stressing over code quality; and using branches and worktrees.
