Module 0: Start here
The one rule
This course has many tips and one rule. The tips make you faster. The rule is what makes you a developer instead of someone who copies and pastes. If you remember only one thing from this course, make it this.
The rule
Never ship code you cannot explain.
"Ship" means to put something where other people will use it: publish a website, submit an assignment, send code to a client, push to GitHub.
"Explain" does not mean you could have written every line yourself from memory. It means that if someone points at any part of your code and asks "what does this do?", you can give a sensible answer in your own words.
If the AI hands you something you do not follow, ask it to explain before you use it. That is the whole difference between using AI and being carried by it.
Meet your intern
Here is the most useful way to think about AI coding tools.
AI is a fast, confident intern. It has read every book. It has never worked a day.
Imagine a new intern joins your team. They have read millions of pages of code, documentation and tutorials. They type faster than anyone you have met. They never get tired and never get annoyed when you ask the same question five times.
They also have some serious problems.
What the intern is brilliant at
- Writing the boring first version. A whole web page in ten seconds.
- Explaining code you do not understand. Line by line, as many times as you want, in simple English or even Hindi.
- Fixing an error you paste in. Often on the first try.
- Suggesting ideas you had not thought of.
What the intern is terrible at
- Knowing what you meant. It only knows what you said. If you say "make a resume website", it guesses everything else, including your name.
- Admitting it does not know. It almost never says "I am not sure". It will state wrong things with total confidence, in perfect English.
- Checking its own work. Unless you ask, it does not run the code it gave you. It does not open the page and look.
- Knowing what happened recently. Its knowledge stops at a certain date. Tools and websites change after that.
You would not let a brand-new intern send code to a client without someone reading it first. That someone is you.
Why this rule matters so much
Because the code will break
Every project breaks at some point. A button stops working. A page goes blank. If you understand your code, you can usually find the problem in minutes. If you do not, you are stuck pasting errors into a chat and hoping. Sometimes the AI "fixes" one thing and breaks two others, and you cannot tell.
Because of interviews
Interviewers love to point at a project on your resume and say "walk me through this". If you built it with AI and understand it, that is a great conversation. If you cannot explain it, it falls apart in the first two minutes, and the interviewer remembers.
Because of safety
Code can leak passwords, expose people's data, or cost money. AI tools sometimes write code with these problems. If you cannot read what you are shipping, you cannot catch them. Module 12 covers this properly.
Is using AI cheating?
No. Professional developers use AI tools every day. Using a tool well is a skill.
What is dishonest is passing off work as a skill you do not have. Saying "I built this" is fine. Saying "I know React" when you only know how to paste prompts is not, and it will catch up with you.
Whenever a college assignment or a job has its own rules about AI, follow them. When in doubt, ask, and say clearly how you used AI.
What the rule looks like in practice
You do not need to understand everything at the same depth. Here is a realistic standard for a beginner.
| You should be able to... | Example |
|---|---|
| Say what each part of the file does | "This part is the heading, this part sets the colours, this part handles the button click." |
| Explain any line someone points at, roughly | "This line saves the expense to the browser's storage." |
| Know what you would change to change something | "To change the button colour, I would edit this line in the CSS." |
| Say what you do not know yet | "I do not fully understand this regular expression yet. It checks the email format." |
That last row matters. "I do not fully understand this yet, and here is what I know about it" is an honest, professional answer. Pretending is not.
How to follow the rule
It takes about five extra minutes per change. Every time the AI gives you code:
- Read it once, top to bottom, even if you only understand half.
- Ask about anything unclear. Do not skip the scary-looking parts.
- Run it and look. Does it actually do what you wanted?
- Say it back. Explain it in your own words, to the AI, a friend, or yourself.
Here is a prompt you will use again and again. Copy it and keep it handy:
Explain this code to me line by line. I am a beginner. For each part, tell me what it does and why it is needed. Use simple English and no jargon. If a line uses a word I may not know, define it.
[paste the code here]And here is one that checks whether you really understood:
I am going to explain what this code does in my own words. Tell me what I got right, what I got wrong, and what I missed. Be honest, not polite.
Code:
[paste the code here]
My explanation:
[write your explanation here]Module 4 has a whole lesson on making AI teach you: Making AI explain.
The rule, one more time
Use AI as much as you like. Let it write the first draft, fix errors, and suggest ideas. But before anything goes out with your name on it, make sure you could explain it.
Fast is good. Fast and understood is what gets you hired.
Your task
- Open Gemini at gemini.google.com and sign in with your Google account. If you do not have one yet, read along now and do this after the Accounts lesson.
- Copy and send this prompt:
You are a patient teacher. Explain what HTML is to someone who has never coded, in four sentences, with no jargon.- Read the answer. Then explain HTML in your own words, out loud or in writing, without looking at the answer.
Check yourself
Enjoying this? This is one of the free lessons. The full course has 78 lessons and three projects you publish yourself.