After the Gaokao, a New Starting Line in the AI Era

This is a text edition. Images, polls and embedded content from the original are not yet included; see the original source below.

The two months after the gaokao are quietly creating a new gap.

After the admission letter a new gap is only beginning.

In the same summer, some simply wait for the semester to start; others have already made their first university project.

What is worth preparing in advance is not another round of textbook memorization, but learning to turn an idea in your head into something runnable, demonstrable, and improvable with AI.

For gaokao graduates and incoming university students; beginners can start too.

Imagine a scene after the semester begins.

A student club needs an introduction page; a course group wants to present its research findings; you want a more distinctive personal homepage. Someone naturally opens an AI tool, breaks the idea down into requirements, generates, revises, and checks it step by step, then shares a link anyone can visit.

Others still react first: “But I am not a computer science major. I cannot code.”

The unsettling part is not that others have learned another programming language. It is that they have begun using AI to amplify their ideas while you still treat it only as a search box and chatbot.

At first the gap may amount to just a web page or a club presentation. As it accumulates, it becomes project experience, expressive ability, approaches to solving problems, and speed of action when facing new tools.

This is not a plea to turn the summer after the gaokao into another cram school. Quite the opposite: you only need one sufficiently small project to go through the entire process from idea to finished work once.

01 / WHY NOW

Do not memorize syntax first; complete one creative project first.

Traditional introductions to programming often begin with variables, loops, and functions. That is valuable, but for a beginner who simply wants to find out whether AI can help make an idea real, the starting path can be more intuitive.

Vibe Coding does not mean having AI do everything in one click, nor does it mean ignoring the results. It is more like a collaboration:

IDEA → INTERFACE

You put forward the idea, provide context, and judge whether the result meets your expectations.

The AI Agent helps plan, generate code, call tools, and make changes.

Finally, you inspect the page, identify problems, iterate, and complete acceptance checks.

The point of this lesson is not to memorize a software button, but to establish a starting method you can use again.

02 / 150 MINUTES

One lesson, six stages

00:00–00:20 · 20 MIN

What are AI and Vibe Coding?

Understand intuitively what large language models and generative AI are, and what AI is and is not good at.

00:20–00:40 · 20 MIN

Models, APIs, and Chatbox

Understand the relationships among models, APIs, API keys, requests, and responses.

00:40–01:00 · 20 MIN

Agents, prompts, and workflows

Learn about agents’ planning and tool-calling abilities, and organize prompts around goals, constraints, and acceptance criteria.

01:00–01:30 · 30 MIN

Break down and modify a classic project

Begin with a personal homepage, to-do list, or information card and personally edit its content, styling, and basic interactions.

01:30–02:00 · 30 MIN

Create your own small project

Define the target users, core function, page structure, and visual direction, then generate a first version.

02:00–02:30 · 30 MIN

Polish, debug, and publish

Check mobile responsiveness, links, and the console; publish the project as an accessible web page if conditions allow.

idea-to-interface.workspace

PROMPT

Goal: make a clear, welcoming personal project page.

Constraints: single-page HTML, mobile-friendly, no horizontal overflow.

Acceptance: links work, no console errors, page is accessible.

● preview ready · responsive · accessible

03 / THE WORKFLOW

Not one-shot generation but six rounds of iteration

Propose an idea: first explain what problem you want to solve.

Clarify requirements: specify users, features, pages, and project scope.

Write a prompt: add context, constraints, and acceptance requirements.

Generate a first version: have AI produce the initial project and then understand the key files.

Debug and iterate: change one small thing at a time and check the result each time.

Publish the project: complete acceptance checks and, if possible, upload it as a web page.

04 / TAKEAWAYS

You leave with more than a web page

A runnable small project

A framework for breaking down requirements

A method for iterating prompts

A debugging and acceptance checklist

Source code and a published address

A path for further learning

The realistic limits of one lesson

150 minutes is suitable for a web page or small tool with controlled scope; it does not promise a large app or commercial system in one lesson. Going live during the session also depends on the network, accounts, third-party platforms, and project size. Even without publishing, you will retain the code, debugging results, and clear next steps.

05 / INSTRUCTOR

Kerry Chia

A student of Data Computing and Applications (Data Intelligence) at Renmin University of China. Teaching interests include Prompt Engineering, AI tools, and introductions to Python and C/C++.

I also built my personal homepage gradually with an AI Agent, from requirements and design to code and deployment. You can view it as a real example:

profile.kerrychia.site

The full course description, schedule, and FAQ are here:

course.kerrychia.site

Click Read Original at the end to jump there.

START SMALL · BUILD FOR REAL

Turn an idea in your head into your first accessible project.

698 CNY / 2.5h

Add WeChat kerrychia Note: Vibe Coding course