GH-300 domain - 13% of the exam

Understand GitHub Copilot data and architecture

Understand GitHub Copilot data and architecture is 13% of the GitHub Copilot (GH-300) exam. These are the objectives it covers, each with practice questions and worked explanations.

Objectives in this domain

Sample question from this domain

Free sampleUnderstand GitHub Copilot data and architecturemedium

A developer typing in their editor with GitHub Copilot wants to understand what travels to the Copilot service when a ghost-text completion is requested. Which description of the data flow for an inline completion request is accurate?

  • AThe editor uploads the developer's entire local repository to GitHub on every keystroke so the model can index the whole project before replying.
  • BThe plugin compiles and executes the project locally, then transmits only the runtime output of the program to the Copilot proxy for completion.
  • CThe request is answered entirely on the developer's machine by a bundled offline model, so no prompt data leaves the editor at any point.
  • DThe editor builds a prompt from nearby code and open-file context, sends it through the Copilot proxy to the model, and returns a suggestion to the editor. Correct
Describe how an inline completion sends a context-built prompt through the Copilot proxy to a cloud model and returns a suggestion. For an inline completion the editor extension assembles a prompt from nearby code and open-file context, sends it through the Copilot proxy to the cloud-hosted model, and returns the suggestion; it neither uploads the whole repository per keystroke nor runs an offline model locally.

Why A is wrong: Copilot does not upload the whole repository on each keystroke; it assembles a bounded prompt of surrounding context, so describing a full-repo upload per keystroke overstates what is transmitted.

Why B is wrong: Copilot does not compile or run the project; it works from source text context rather than runtime output, so this confuses execution with the contextual prompt that is actually sent.

Why C is wrong: Inline suggestions are produced by a cloud-hosted model reached through the Copilot service, not a bundled offline model, so claiming nothing leaves the editor is incorrect.

Why D is correct: An inline request packages surrounding code and open-file context into a prompt, routes it through the Copilot proxy to the model, and returns the generated suggestion, which is the documented completion flow.

Other domains in this exam

See also the GH-300 cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by GitHub. Original, blueprint-aligned practice material only.