| トピック | 出題範囲 |
| トピック 1 | - How GitHub Copilot Works and Handles Data: Designed for Machine Learning Engineers and Data Privacy Specialists, this section covers the data lifecycle and processing behind Copilot’s code suggestions. It explains how context is gathered, prompts constructed, responses generated, and post-processed through proxy services. Candidates understand Copilot’s data policies, handling of inputs, and limitations such as context window size and data age influencing suggestion relevance.
|
| トピック 2 | - Domain 6: Testing with GitHub Copilot: This section measures abilities of QA Engineers and Test Automation Specialists to use Copilot for test generation, including unit and integration tests. It explains how Copilot can identify edge cases and assist in writing assertions. The domain also covers different Copilot subscription SKUs, privacy considerations, organizational code suggestion settings, and configuration files related to Copilot.
|
| トピック 3 | - GitHub Copilot Plans and Feature: This domain targets Product Managers and DevOps Engineers and focuses on understanding the various GitHub Copilot subscription plans like Individual, Business, and Enterprise, including distinctions and management features. It covers how Copilot is integrated into IDEs, different triggering methods for code suggestions, organizational policy management, subscription administration via API, and effective use of Copilot Chat and Knowledge Bases. Candidates also learn about CLI usage and configuration.
|
| トピック 4 | - Responsible :This section of the exam measures skills of AI Ethics Officers and Risk Managers and covers the responsible and ethical usage of AI technologies. It explains the risks and limitations associated with generative AI tools, including biases in training data and the need to validate AI outputs. Candidates learn how to operate AI responsibly by identifying potential harms such as bias, fairness, privacy concerns, and mitigating these harms by applying ethical AI principles.
|
| トピック 5 | - Domain 4: Prompt Crafting and Prompt Engineering This section measures skills of Software Developers and AI Interaction Designers in effectively crafting prompts to optimize Copilot’s output. It reviews foundational concepts such as prompt components, the role of language in prompting, zero-shot vs. few-shot prompting, and how chat history influences responses. Best practices and engineering principles for prompt design and training methods are also covered.
|
| トピック 6 | - Developer Use Cases for AI: Targeting Software Engineers and Technical Leads, this domain elaborates on how AI improves developer productivity across common tasks like learning new languages, translation, documentation, debugging, data science, and refactoring. It discusses Copilot’s support in software development lifecycle management and highlights its limitations. Use of the productivity API to track Copilot’s impact is also included.
|