Steps to Build LLM Coding Intuition

1. Define Your Goal
   Clearly state what you want to build or solve.
   Example: “Create a REST API endpoint for user registration.”

2. Break Down the Problem
   Divide the goal into smaller, manageable tasks.
   Example: “Set up routing, handle input validation, connect to the database, write tests.”

3. Ask for Best Practices
   Prompt the LLM for recommended patterns, libraries, or approaches.
   Example: “What’s the best way to structure a Go REST API project?”

4. Start with Explicit Prompts
   Request code, explanations, or templates for each sub-task.
   Example: “Generate a handler function for user registration in Go.”

5. Review and Understand
   Read the generated code, ask for explanations, and make sure you understand each part.
   Example: “Explain how this handler validates input.”

6. Adapt and Experiment
   Modify the code to fit your needs, and try alternative prompts to see different solutions.
   Example: “How would this look using a different validation library?”

7. Test and Validate
   Ask the LLM to generate tests, then run and review them.
   Example: “Write a unit test for the registration handler.”

8. Refactor and Optimize
   Request suggestions for improvements or optimizations.
   Example: “Suggest improvements for error handling in this function.”

9. Document and Comment
   Use the LLM to help document your code and decisions.
   Example: “Write a docstring for this function.”

10. Reflect and Iterate
    After completing the task, reflect on what worked, what didn’t, and how the LLM helped.
    Example: “What could be improved in this workflow?”

---

Summary Rule:

• Start by following best practices and explicit prompts (“play by the book”).
• As you gain confidence, experiment, adapt, and trust your own judgment (“play by your heart”).