AI development tools can be "resistant," ignoring change requests. A powerful technique is to prompt the AI to consider multiple options and ask for your choice before building. This prevents it from making incorrect unilateral decisions, such as applying a navigation change to the entire site by mistake.

When asked to modify or rewrite functionality, LLMs often attempt to preserve compatibility with previous versions, even on greenfield projects. This defensive behavior can lead to overly complex code and technical debt. Developers must explicitly state that backward compatibility is not a requirement.

When using AI development tools, first leverage their "planning" mode. The AI may correctly identify code to change but misinterpret the strategic goal. Correct the AI's plan (e.g., from a global change to a user-specific one) before implementation to avoid rework.

Before writing any code for a complex feature or bug fix, delegate the initial discovery phase to an AI. Task it with researching the current state of the codebase to understand existing logic and potential challenges. This front-loads research and leads to a more informed, efficient approach.

When using AI for complex but solved problems (like user permissions), don't jump straight to code generation. First, use the AI as a research assistant to find the established architectural patterns used by major companies. This ensures you're building on a proven foundation rather than a novel, flawed solution.

Use 'stop hooks' in Claude Code to create an automated quality gate. After code generation, the hook runs checks like type checking or linting. If errors exist, the output is fed back to the AI with a prompt to fix them, creating a self-correcting workflow.

When building with rapidly evolving LLMs, avoid creating rigid structures or "scaffolding" to compensate for current model weaknesses. This technical debt becomes a liability when more capable models emerge. Instead, design systems that can leverage future improvements without a complete rebuild.

Before letting an AI modify files, use its 'plan mode' to have it outline its approach. Ask it to identify files to change, potential risks, and what it's intentionally leaving out. This provides a chance to review and course-correct, preventing wasted effort and unexpected side effects, similar to a human code review process.

When an AI coding assistant asks you to perform a manual task like checking its output, don't just comply. Instead, teach it the commands and tools (like Playwright or linters) to perform those checks itself. This creates more robust, self-correcting automation loops and increases the agent's autonomy.