GitHub · GH-300
How Copilot works with your data
GH-300 · Lesson 7 of 11 · 41:02
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Prepare for the GH-300 exam (GitHub Copilot certification). In this lesson: how a request travels: the context Copilot gathers (code around the cursor, file names, open tabs, project structure) and fill in the middle, the proxy in GitHub's Azure tenant, toxicity filters, the model, post-processing with security checks and the public code filter, the suggestion lifecycle, data retention for editor suggestions and for chat, and the limits of large language models. Includes three original practice questions.
Preview the topics
- Where are attempts to manipulate Copilot, into revealing how it works, blocked?
- A generated response fails the security checks during post-processing. What happens to it?
- In a very large repository, Copilot ignores a helper function defined in a file that isn’t open. What’s the most likely reason?