AI holds tremendous potential and moves ahead at breakneck speed. One of the areas in
which it shines is helping you increase your productivity as a developer, in and out of
Visual Studio. In this deep dive course, which starts where the related getting started
course stopped, you�ll learn how to do that. The course is focused on GitHub Copilot, but
other assistants and approaches are also covered. You�ll learn how to use it in a variety
of IDEs. You�ll learn how it can help write tests, optimise code, commit code, review PRs,
and even iteratively help you implement a user story with Copilot Edits. After that, a
substantial part of the course is dedicated to evaluating LLMs: you�ll get a taste of Mistral
Large, Claude Sonnet, GPT4o and others. A variety of LLMs is then used to teach you advanced
prompt engineering techniques to improve the results you�ll get. The final section of the
course is about extension and AI agents: you�ll learn how to write your own agent that integrates with GitHub Copilot.
Course Curriculum
4h 55m 9 sections
Welcome
01:21Free preview
What will you learn in this course?
02:27Free preview
Who is the course for and prerequisites
01:14Free preview
Introduction
00:40
Positioning GitHub Copilot
04:12
Setting Up GitHub Copilot in VS Code
01:45
Working With GitHub Copilot in VS Code
09:46
Setting Up GitHub Copilot in JetBrains Rider
02:13
Working With GitHub Copilot in JetBrains Rider
04:10
Working With GitHub Copilot in the GitHub CLI
07:20
Working With GitHub Copilot in Other IDEs
01:32
Section Recap
02:20
Introduction
01:13
Getting to Know the Demo Codebase
04:15
Unit Tests, Integration Tests and End-to-End Tests
03:24
Generating Dummy Data
13:58
Generating and Running Unit Tests
10:52
Generating and Running Integration Tests
15:43
WebApplicationFactory: Integration Tests or End-to-End Tests?
03:51
Generating and Running End-to-End Tests
08:03
Generating Test Requests
01:27
Section Recap
00:49
Introduction
00:46
Improving Your Codebase
09:23
Working With Copilot Edits
07:58
Section Recap
01:01
Introduction
00:32
Committing Code and Creating Pull Requests in Visual Studio
05:11
Creating Pull Requests on GitHub
06:54
Section Recap
01:04
Introduction
00:42
LLMs, From Generic to Specific
08:14
How LLMs Are Rated
10:16
Should You Be Running Benchmarks Yourself?
01:15
Comparing Popular LLMs
05:45
Evaluating ChatGPT (GPT4o) for Development
06:46
Evaluating Mistral (Mistral Large) for Development
06:04
Evaluating Claude (Sonnet) for Development
06:25
Integrating Claude in Your IDE
06:41
Evaluating Amazon Q Developer for Development
09:54
Integrating With LLMs From Code
02:05
Section Recap
02:37
Introduction
00:37
Prompts, Prompt Engineering and the Lingo Problem
04:41
Manipulating LLM System Prompts
04:46
Zero-, One-, and Few-shot Prompting
06:23
Weighted Prompting
03:08
Chain-of-Thought (CoT) Prompting
04:03
Reasoning and Acting (ReAct) Prompting
04:39
Retrieval-augmented Generation (RAG) Prompting
02:03
Section Recap
03:23
Introduction
01:18
About AI Agents and Extensions
01:58
Discovering Extensions
02:37
Installing and Using an Extension
02:39
Creating an AI Agent with Retrieval-augmented Generation (RAG) Prompting Support
04:28
Creating a Custom AI Agent: Plumbing
15:26
Creating a Custom AI Agent: Implementation
19:08
Creating a Custom AI Agent: Tightening the Implementation
Kevin is a freelance solution architect, author & consultant, living in Antwerp (Belgium). He started working in the IT sector over 20 years ago, and is an 11-time Microsoft MVP. He's focused on architecture & security for web applications & integration components, using .NET and Azure. He's a keen proponent of open-source software. Also: wine.