AI-Assisted Engineering

AI is the copilot. Engineering is your responsibility.

Students learn to use modern AI development tools responsibly, from Week 01 onward — and to explain and defend every line of code they submit, AI-assisted or not.

The workflow

From requirement to deployed code

Requirement

↓

Break down the problem

↓

AI-assisted planning

↓

Generate / modify code

↓

Developer reviews code

↓

Run tests

↓

Debug

↓

Refactor

↓

GitHub Pull Request

↓

Code Review

↓

Deploy

Modern AI development workflow

What students actually practice

AI-assisted coding

Generate boilerplate
Explain unfamiliar code
Refactor
Create tests
Debug
Documentation

AI-assisted debugging

Provide logs
Analyse errors
Reproduce issues
Evaluate proposed fixes
Verify the solution

AI code review

The questions students learn to ask

Is this secure?

Is this maintainable?

Is this performant?

Are there edge cases?

Are tests missing?

Is the architecture appropriate?

AI limitations

AI is not marketed as magic here

Students are taught to recognise where AI-generated code commonly goes wrong, before it ships:

Hallucinated APIs Incorrect code Security vulnerabilities Outdated information Over-engineering Dependency mistakes Poor architecture

Tools

Tools students learn to use

Claude, Cursor, ChatGPT and GitHub Copilot are taught as tools within the workflow above — Zerobug is not affiliated with, sponsored by, or endorsed by any of these companies.