Let CI Be The Referee

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-coder-ci-cd-agent-creators

1. Why should a coding agent's confident tone not be taken as a reliable indicator of correct code?

   1. Confident agents are more likely to pass code review without scrutiny
   2. Confidence is a personality trait programmed into the model, not an indicator of output quality
   3. Confident responses indicate the agent has successfully compiled all external documentation
   4. Confidence level in AI responses correlates strongly with actual code correctness

2. When setting up work for a coding agent, what should be identified first, before selecting which tool to use?

   1. The exact API endpoints the agent needs to call
   2. The job or task that needs to be accomplished
   3. The specific code editor the agent will operate in
   4. The version of the programming language required

3. What is the recommended scope when assigning work to a coding agent?

   1. A comprehensive rewrite of the entire codebase to ensure consistency
   2. The smallest useful piece of work that delivers value when completed
   3. A single file modification to keep changes simple
   4. The largest possible feature set to maximize the agent's value

4. How should you run and test the output produced by a coding agent?

   1. As a fan of the tool, focusing on what it did well
   2. As a skeptical user who wants the feature to actually work
   3. As a developer reviewing the syntax and style only
   4. As a manager checking if the deadline was met

5. Before sharing an AI agent's code contribution with a team, what three things should be inspected?

   1. The diff, data access patterns, and failure paths
   2. The color scheme, UI layout, and user experience
   3. The file names, variable names, and comment style
   4. The git history, commit messages, and branch name

6. What role does CI play when a coding agent is integrated into a development workflow?

   1. CI automatically rewrites any code that doesn't meet style guidelines
   2. CI serves as the primary interface between developers and the AI agent
   3. CI acts as a neutral referee that validates whether the agent's changes are safe
   4. CI provides creative suggestions for improving the code architecture

7. A team wants to get value from coding agents. What is the recommended approach for integrating them into existing workflows?

   1. Run agents in isolation on separate infrastructure from the main codebase
   2. Have agents operate only on documentation and never touch production code
   3. Use agents on an ad-hoc basis whenever a developer gets stuck
   4. Wire agents into existing CI checks so their output is automatically validated

8. What question should you ask to determine what data an application should never expose?

   1. What data does the user explicitly request in the interface?
   2. What data would cause competitive advantage if competitors saw it?
   3. What data should the app or agent never expose?
   4. What data would make the application run faster if exposed?

9. What question identifies the rollback strategy if an AI agent's output causes problems in production?

   1. Who should be blamed if the code fails?
   2. How long should the team wait before deploying a fix?
   3. Should the agent be used for future projects?
   4. What rollback path exists if the output is wrong?

10. According to the principles discussed, what proves that an AI agent's code change actually works?

    1. A manager's approval of the changes
    2. The agent's confidence level in its response
    3. A test that verifies the expected behavior succeeds
    4. The fact that the code compiles without errors

11. Why is turning an AI demo into something 'observable' an important skill?

    1. Observability reduces the amount of code that needs to be written
    2. Observability allows the team to understand what the code is doing and detect problems
    3. Observability makes the code appear more professional to stakeholders
    4. Observability makes the code run faster in production

12. What makes code 'reversible' and why is this important when using AI agents?

    1. Reversible code can be undone by pressing Ctrl+Z in the editor
    2. Reversible means having a clear way to roll back changes if something goes wrong
    3. Reversible means the agent can undo its own mistakes automatically
    4. Reversible code can run on both Windows and Linux operating systems

13. What does it mean for AI-generated code to be 'safe enough for another person to use'?

    1. The code has passed a security scan and been approved by legal
    2. The code is well-documented with comments explaining every function
    3. The code runs faster than human-written alternatives
    4. The code has appropriate safeguards, data protection, and error handling so it won't harm users or systems

14. The lesson mentions that AI can quickly create a working demo. What distinguishes a real developer skill from simply generating demos?

    1. The ability to turn demos into observable, reversible, and safe code for production use
    2. The ability to prompt the AI more effectively than other developers
    3. The ability to reduce the amount of code the AI writes
    4. The ability to write more complex code than the AI can generate

15. Which four automated checks are typically performed by a CI pipeline to validate code?

    1. Code review, pair programming, sprint planning, and release notes
    2. Documentation generation, security scanning, database migration, and deployment
    3. Code formatting, color scheme validation, performance profiling, and user acceptance testing
    4. Lint checks, type checking, running test suites, and building the application