Who this course is for
Developers who already write code for a living and want to use AI assistants without producing more bugs, security holes or code nobody understands.
Examples use C# and .NET (ASP.NET Core, EF Core, xUnit), but every technique works in Java, Python, JavaScript or any other language. If you are new to AI assistants in general, Prompt Writing That Works is a good companion.
What you will learn
- When to use a chat assistant, an in-editor assistant or an agent that edits files for you
- How to write prompts for code: context, constraints and examples
- How to get up to speed on an unfamiliar codebase quickly
- How to go from a stack trace to a root cause and a fix
- How to write unit tests with AI, and how to review them properly
- How to refactor in small, safe steps
- How to write and tune SQL and EF Core queries with AI help
- What never to paste, and what to always review: secrets, security and licensing
What you need
- Your usual IDE (Visual Studio, VS Code, Rider or similar) and a project you can experiment on
- At least one AI tool: a chat assistant (ChatGPT, Claude, Gemini) and, if your company allows it, an in-editor assistant such as GitHub Copilot
- Source control (git), so every AI change can be reviewed and undone
One rule for the whole course
You own every line that ships. AI writes fast, but it does not carry the pager. Review, test and understand everything before you commit it.
Click Mark as complete at the end of each lesson to track your progress.
Course content
- 1 Chat Assistants, In-Editor Assistants and Coding Agents Understand the three kinds of AI coding tools, what each is good at, and how to combine them in a normal working day. 10 min
- 2 Writing Prompts for Code: Context, Constraints, Examples The details that turn generic AI code into code that fits your project: versions, existing types, conventions, constraints and a worked example. 12 min
- 3 Understanding an Unfamiliar Codebase With AI Get productive in a new or legacy codebase faster: map the structure, trace a request end to end, and turn what you learn into notes for the team. 11 min
- 4 Debugging: From Stack Trace to Root Cause and Fix Give AI the right evidence (error, stack trace, code, what changed) to find root causes faster, and avoid fixes that only hide the symptom. 13 min
- 5 Writing Unit Tests With AI, and Reviewing Them Properly Use AI to list test cases and write xUnit tests quickly, then review them so they actually catch bugs instead of just raising coverage. 13 min
- 6 Refactoring Safely in Small Steps Use AI to plan and carry out refactors as a series of small, behaviour-preserving steps, each one tested and committed, instead of one risky rewrite. 12 min
- 7 SQL and EF Core Queries With AI Help Get correct SQL and LINQ by giving AI your schema, check the SQL EF Core actually generates, and fix common performance problems like N+1 queries. 13 min
- 8 Security and Licensing: What to Never Paste, What to Review Protect secrets and company code, catch the security mistakes AI often makes, check that suggested packages are real, and respect code licences. 13 min