Getting Started with AI Coding Tools: A Student's Guide

Alex Rivera

CS Student & Developer

February 10, 20253 min read

AI code generators have created a split in the CS student community. Some see them as cheating. Others use them as an unfair advantage. The truth is more nuanced — and far more interesting.

Used correctly, AI coding tools can dramatically accelerate your learning. Used incorrectly, they'll leave you unable to code without a crutch.

Here's how to use them the right way.

Understanding What AI Code Tools Actually Do

AI code generators (like those in Geleza) don't "know" how to code in the human sense. They're pattern-matching engines trained on billions of lines of code. They excel at:

Common patterns and algorithms
Boilerplate code
Explaining what code does
Converting between languages
Suggesting completions

They're weaker at:

Novel architecture decisions
Debugging complex logic errors
Code that requires deep domain knowledge

If you copy AI-generated code without understanding it, you're borrowing against future confusion. The debt comes due during exams and job interviews.

The Right Mental Model: AI as a Rubber Duck

Programmers use a technique called "rubber duck debugging" — explaining your problem to an inanimate object forces you to think through it clearly. AI is a rubber duck that talks back.

Use it to:

Explain a concept you don't understand
Review code you've already written
Generate examples of patterns you're learning
Ask "why" questions, not just "what" questions

Practical Workflow for Students

Here's a workflow that builds skills while leveraging AI:

1. Attempt the problem yourself first (30 minutes minimum)
2. If stuck, describe the problem to AI in plain English
3. Ask for hints, not solutions: "What approach should I take?"
4. If you get code: read every line, understand each one
5. Rewrite it from scratch without looking
6. Test edge cases manually

Learning New Concepts with AI

AI is exceptional for learning new syntax and concepts:

python

# Ask: "Show me list comprehension in Python with a real example"
# Result:
squares = [x**2 for x in range(10) if x % 2 == 0]
# vs the for-loop equivalent:
squares = []
for x in range(10):
    if x % 2 == 0:
        squares.append(x**2)

The AI shows you both versions, helping you understand the why behind new syntax.

Questions to Ask AI (That Actually Build Skills)

Instead of: "Write me a sorting algorithm"

Ask:

"Explain why bubble sort is O(n²) in the worst case"
"What's the difference between merge sort and quicksort for nearly-sorted data?"
"Show me how recursion works in merge sort step by step"

The quality of your learning scales with the quality of your questions. Treat AI like a knowledgeable tutor, not an answer machine.

Red Flags: When AI is Hurting Your Learning

Watch for these warning signs:

You can't explain code you "wrote" with AI help
You immediately go to AI before attempting a problem
You accept the first solution without questioning it
Your solutions work but you don't know why

If you notice these patterns, implement a "no AI for 48 hours" rule periodically to rebuild your independent problem-solving muscles.

The Bottom Line

AI coding tools are like calculators. A student who uses a calculator without understanding arithmetic will struggle. A student who understands arithmetic and uses a calculator is faster and more capable.

Learn to code. Use AI to accelerate that learning. Never let it replace the thinking.

Found this helpful?

Share it with your classmates.