8.2. Coding with Prompts

🔹 Basic Code Generation Prompt: “Write a Python function to check if a number is a palindrome.” Output:

def is_palindrome(n):
    return str(n) == str(n)[::-1]

✅ Use precise language: "function", "Python", "return Boolean".

In this section, we learn how to generate basic code using prompts. A prompt in this context is a specific instruction we give an AI to create a piece of code. For example, asking the AI to write a Python function that checks if a number is a palindrome means we're instructing it to come up with a code that can identify whether a number reads the same forwards and backwards. The output provided shows the code itself, which defines a function is_palindrome. This function converts the number n into a string and checks if it is equal to its reverse (str(n)[::-1]). It's important to use precise language in the prompt because it helps the AI understand exactly what you are asking for.

Think of this like asking a chef to make a specific dish. If you just say, 'make something with chicken', the chef might end up making a soup, a salad, or a roast. But if you say, 'make a chicken salad', the chef knows exactly what to do. This is similar to how precise language in prompts helps the AI produce the exact code you need.

🔹 Multistep Instruction Prompt: “Write a Python script that reads a CSV file, filters rows where age > 30, and saves the output.” Use delimiters for long code:

# script goes here

This chunk explores how to give multistep instructions to an AI to perform a specific coding task. In the example, the prompt asks the AI to create a Python script capable of reading data from a CSV file, filtering the rows based on certain criteria (in this case, where the age is greater than 30), and then saving the filtered data somewhere. Using delimiters helps to clearly separate the code from the rest of the prompt, making it more readable and reducing potential misinterpretations of the instructions. Each step can be seen as a building block toward completing the overall task.

Imagine you're assembling furniture from a store. If the instructions you get include a detailed series of steps – like 'first attach the legs, then place the tabletop on top' – you can follow along easily. On the other hand, if the instructions aren't clear or if parts of it are mixed in with irrelevant information, you might end up assembling something incorrectly. The same principle applies to coding instructions with AI.

🔹 Debugging and Fixes Prompt: “Fix this error in JavaScript code:

let x = [1, 2, 3];
console.log(x[3].toUpperCase());

Explain the error.” Output: “x[3] is undefined, so calling toUpperCase() throws an error.”

This section deals with debugging code using prompts. When providing a prompt to the AI, the user presents a piece of code with an error and asks the AI to identify and explain the error. The example shows JavaScript code that attempts to access an index in an array that does not exist (x[3]), resulting in an 'undefined' value, which causes an error when the method toUpperCase() is called on it. The AI explains that the index is out of range for the array defined, which has only three elements indexed from 0 to 2.

Imagine you're trying to find a specific book on a shelf, but the shelf only has two books. If you reach for the third book that isn’t there, you’re going to be confused and frustrated. Similarly, the code tries to 'reach' for something that doesn't exist, causing an error. Debugging helps identify these missteps just like asking a librarian could clarify which books are actually available.

--