Quest 1 - First AI Completion

Quest 1: First AI Completion

easy 25 minutes

---

🎯 Learning Objectives

• ✅ How to use AI coding tools for code generation
• ✅ Why you must verify AI output before trusting it
• ✅ How edge cases can silently break naive AI suggestions
• ✅ The practice of reading failing tests to understand requirements

---

📖 Concept: AI Code Completion

AI coding tools like GitHub Copilot, Cursor, and Claude Code can generate code from natural language comments. You describe what you want, and the AI writes the code for you.

This is incredibly powerful — but it comes with a critical caveat: AI output is a starting point, not a finished product. The AI generates code that looks correct for the obvious cases, but it often misses edge cases that a human developer would catch.

Think of AI code completion like a junior developer who writes code fast but doesn’t think about edge cases. Your job is to be the senior developer who reviews, tests, and validates.

---

⚙️ How It Works

The AI Code Completion Workflow

1. Write a comment describing what you want
         ↓
2. AI generates code suggestion
         ↓
3. Accept the suggestion
         ↓
4. Run tests — READ the failures carefully
         ↓
5. Fix edge cases the AI missed
         ↓
6. Verify all tests pass

---

🧩 Interactive Demo

Loading visualization...

AI Code Completion Flow

---

Why Verification Matters

Consider this AI suggestion for a factorial function:

// Naive AI suggestion
function factorial(n) {
  if (n <= 1) return 1;
  return n * factorial(n - 1);
}

This looks correct, right? Let’s check:

Input	Expected	AI Output	Correct?
factorial(5)	120	120	✅
factorial(1)	1	1	✅
factorial(0)	1	1	✅
factorial(-1)	❌ Error	1	❌ WRONG

The AI’s if (n <= 1) return 1 returns 1 for negative numbers, but factorial of a negative number is undefined. The function should throw an error.

This is the #1 mistake developers make with AI: accepting the first suggestion without testing edge cases.

---

💡 Example: The Right Way

Here’s how an experienced developer handles AI code completion:

Step 1: Write a descriptive comment

// Calculate factorial of n using recursion.
// Throw error for negative numbers (undefined).
// Throw error for non-integers.

Step 2: Review the AI suggestion critically

function factorial(n) {
  if (typeof n !== 'number' || !Number.isInteger(n)) {
    throw new Error('Input must be an integer');
  }
  if (n < 0) {
    throw new Error('Factorial is undefined for negative numbers');
  }
  if (n <= 1) return 1;
  return n * factorial(n - 1);
}

Step 3: Run the tests and READ the output

node test.js

Don’t just glance at “all tests pass” — read each test case to understand what it’s checking.

Step 4: If tests fail, understand WHY

The failing test will tell you exactly what edge case you missed. This is valuable feedback, not a nuisance.

---

⚠️ Common Mistakes

Mistake 1: Trusting AI blindly

“Copilot suggested it, so it must be correct” → Always run tests. Always check edge cases.

Mistake 2: Not reading test failures

“Tests failed, let me just tweak until they pass” → Read the failure message. Understand what the test expects.

Mistake 3: Copying from solution files

“I’ll just peek at the solution to understand the pattern” → This defeats the learning. Struggle is where growth happens.

Mistake 4: Ignoring error handling

“The happy path works, good enough” → Production code must handle edge cases. Negative inputs, null, undefined, wrong types.

---

📝 Knowledge Check

📝 Knowledge Check

Q1:What is the main lesson of this quest?

AI always generates correct codeAI output is a starting point — you must verify edge casesYou should never use AI for codingTesting is optional when using AI tools

Q2:What is wrong with this naive AI suggestion: `if (n <= 1) return 1`?

It doesn't handle recursion properlyIt returns 1 for negative numbers instead of throwing an errorIt's too slow for large numbersIt only works with positive integers

Q3:When a test fails, what should you do first?

Copy the solution from _solution/solution.jsKeep tweaking code until tests passRead the failure message to understand what the test expectsAsk AI to fix it automatically

---

🏋️ Quest: First AI Completion

Now it’s time to practice! Write a factorial function using AI assistance.

1. Download the starter files:

   npx bluebeltdojo download quest-01-first-completion
   cd quest-01-first-completion

2. Open problem.js in your editor with your AI tool (Copilot, Claude Code, Cursor, etc.)

3. Write a comment describing what you want, then accept the AI’s suggestion

4. Important: Run node test.js and READ the failing tests — especially the negative-input edge case

5. Fix any edge cases the AI missed

6. Verify all tests pass:

   node test.js

7. When all tests pass, submit your solution:

   npx bluebeltdojo submit

💡 Tip: If you’re stuck, re-read the “Common Mistakes” section above. The answer is not in _solution/solution.js — it’s in understanding why your code fails.

---

คำใบ้

• อ่าน instructions ใน problem.js อย่างละเอียด
• ตรวจสอบ edge cases ที่กล่าวถึง (โดยเฉพาะตัวเลขลบ)
• ถ้าติดขัด ลองอ่าน “Common Mistakes” อีกครั้ง — อย่าดู solution โดยตรง