Debugging Event-Driven Systems with AI Help

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-ai-coding-claude-event-driven-debugging-creators

1. In an event-driven system with services A, B, and C communicating via a message broker, what information is essential for AI to accurately stitch together logs from these services?

   1. The programming languages used by each service
   2. The message payload sizes
   3. Correlation IDs and timestamps
   4. The service IP addresses and network topology

2. Which of the following is something AI can reliably do when debugging a distributed event-driven system?

   1. Read the internal state of the message broker directly
   2. Determine which messages are safe to replay without risking duplicate processing
   3. Suggest where missing tracing spans should be added to improve visibility
   4. Access the database transactions behind each event

3. What risk exists when using AI-generated replay scripts for stuck messages in a production system?

   1. The script would fail to compile due to syntax errors
   2. The script might be written in the wrong programming language
   3. The broker would automatically reject the replay attempt
   4. AI could suggest replaying messages that would cause double-charging or other duplicate side effects

4. When AI reconstructs an event timeline from interleaved logs, what is the primary challenge it faces without proper instrumentation?

   1. The logs are stored in different file formats
   2. Without correlation IDs, it cannot reliably determine which entries belong to the same request flow
   3. The logs contain too much detail
   4. The logs are encrypted

5. Why must a human always approve AI-suggested replay scripts before running them in production?

   1. The scripts would violate network security policies
   2. AI does not have permission to access production systems
   3. The scripts are always syntactically incorrect
   4. AI cannot determine which side effects are safe—only a human knows the business logic implications

6. What does the lesson identify as a key limitation of AI in debugging event-driven systems?

   1. AI cannot connect to external APIs
   2. AI cannot read the broker's internal state or understand which messages are safe to replay
   3. AI cannot parse JSON-formatted log entries
   4. AI cannot generate code in any programming language

7. A developer notices that service B should have logged an activity between messages from service A and service C, but no such log exists. How does AI help identify this gap?

   1. By automatically adding fake log entries
   2. By ignoring the gap and assuming everything worked
   3. By deleting the existing logs and starting fresh
   4. By flagging gaps where a service should have logged but didn't, based on the expected timeline

8. What distinguishes debugging event-driven systems from traditional request-response debugging?

   1. Traditional debugging is faster
   2. Event-driven systems do not generate logs
   3. Debugging requires analyzing scattered logs across multiple services with asynchronous communication
   4. Event-driven systems run on different hardware

9. In the context of Kafka, SQS, and Pub/Sub flows, what makes AI useful for debugging?

   1. AI automatically fixes the bugs
   2. These systems store all logs in a single location
   3. AI can process and correlate log entries from multiple services to reconstruct what happened
   4. These systems have no debugging challenges

10. What is a tracing span, and how does AI help with it?

    1. A tracing span is a database table; AI queries it
    2. A tracing span is a log file; AI reads it automatically
    3. A tracing span is a type of error message; AI cannot help with it
    4. A tracing span is an instrumented piece of code that records an operation's execution; AI can suggest where missing spans should be added

11. Why might AI 'confabulate' when debugging event-driven systems?

    1. Because it is given correlation IDs and timestamps
    2. Because the system is too fast
    3. Because the logs are in different languages
    4. Because without correlation IDs, it invents plausible but incorrect connections between unrelated log entries

12. What does the lesson say about using AI to generate replay scripts for stuck messages?

    1. AI can generate useful replay scripts, but they must be approved by a human for idempotency before production use
    2. AI can determine which messages are safe to replay
    3. Replay scripts should never be generated
    4. Replay scripts are automatically safe to run

13. When debugging scattered logs across multiple services, what is the FIRST piece of information you should provide to AI for accurate analysis?

    1. The system architecture diagram
    2. A list of all possible error codes
    3. The complete source code of all services
    4. Correlation IDs that link related requests across services

14. Which statement best describes what AI can and cannot do when debugging event-driven systems?

    1. AI can read broker internals but cannot analyze logs
    2. AI can reconstruct timelines from logs but cannot know which side effects are safe
    3. AI can fix bugs automatically but cannot generate scripts
    4. AI can access databases but cannot parse logs

15. What is the primary value AI provides when debugging event-driven architectures compared to manual analysis?

    1. AI can correlate and synthesize patterns across large volumes of scattered logs that would take humans much longer to piece together
    2. AI can delete old log files to save space
    3. AI can automatically restart failed services
    4. AI can fix the bugs directly in the code