An AI suggestion is a proposed design choice, file structure, code stub, or implementation generated during software development. Treat it as input rather than a decision: apply domain knowledge, retain what fits the project, and reject or revise the rest. section 4.6 section 4.7
Use suggestions productively by keeping the developer in the review loop:
- Ask for requirements, application structure, stubs, or starting code. section 4.7
- Compare outputs from multiple AI tools when different perspectives can improve the result. section 4.6
- Run the generated application, inspect its behavior, and refactor or override suggestions that weaken architecture or maintainability. section 5.3.11
From suggestion to tested Flask code
A coding assistant can provide a starting Flask application with a route returning “Hello, World!”. Running it without errors and checking the resulting web page turns the suggestion into a verified application skeleton, not automatically finished software. section 4.6
from flask import Flask
app = Flask(__name__)
@app.route('/')
def home():
return "Hello, World!"
if __name__ == "__main__":
app.run(debug=True)
Inspect the compact Flask implementation that demonstrates how an AI-generated starting point can become runnable code. section 4.6
