AI-Assisted Build Cache and Bazel Optimization

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-ai-coding-AI-build-cache-optimization-creators

1. An AI tool analyzes a Bazel build profile JSON containing 10,000 cache-miss events. What is the primary advantage of using AI for this analysis compared to manual review?

   1. AI can guarantee a 100% cache hit rate after optimization
   2. AI can automatically group cache misses by rule type and target, identifying patterns faster than human eyeballing of logs
   3. AI can directly measure your team's actual cache hit rate by reading your internal telemetry
   4. AI can modify your BUILD files automatically without developer review

2. Which of the following can an AI determine without access to your build system's telemetry data?

   1. Which BUILD file changes are most likely to improve cache hit rates
   2. Your team's actual cache hit rate
   3. The exact dollar cost of your remote cache usage
   4. The number of remote execution requests made last month

3. What happens when a build target that is NOT hermetic is incorrectly marked as cacheable in Bazel?

   1. Remote execution will be disabled for all targets
   2. The build system will automatically fix the target's dependencies
   3. The cache will be automatically cleared to prevent errors
   4. Phantom passes will occur, causing builds to incorrectly report success when they should fail

4. Before marking a build target as cacheable, what must your team verify first?

   1. That the target has no external dependencies
   2. That the target is hermetic and produces reproducible outputs
   3. That the target runs in under 10 seconds
   4. That the target compiles without errors

5. What is the primary purpose of remote execution in a Bazel build system?

   1. To execute build actions on remote machines instead of the local developer machine
   2. To automatically upload all source code to a cloud repository
   3. To generate build profile JSON files for analysis
   4. To store build outputs in a remote database for backup

6. An AI proposes a change to your BUILD file to improve cache hit rate. What should happen before merging this change?

   1. A hermeticity test should verify the target produces reproducible outputs
   2. The change should be applied only to CI, not local builds
   3. The AI should automatically merge the change to save time
   4. The change should be rejected because AI cannot edit BUILD files

7. Which of these is a key metric for evaluating CI build performance?

   1. Color scheme of the build logs
   2. Cache hit rate
   3. Length of the team's coffee breaks
   4. Number of developers on the team

8. When an AI suggests BUILD file changes 'ranked by expected throughput gain,' what does 'throughput' refer to?

   1. The memory consumption of the build system
   2. The speed at which builds complete, measured in builds per hour
   3. The network bandwidth used during remote execution
   4. The amount of data the build can process per second

9. What type of information can an AI extract from a Bazel profile JSON file?

   1. Your team's internal Slack messages
   2. Timing data for build actions and cache hit/miss events
   3. The entire source code of your project
   4. Which developers wrote which code files

10. What is a 'build cache' in the context of Bazel?

    1. A documentation system for build best practices
    2. A local database that stores compiled binaries for reuse across builds
    3. A network folder where developers share source code
    4. A tool that measures code coverage

11. Why might AI suggest changes to your remote-cache configuration?

    1. To increase the likelihood of cache contamination
    2. To improve cache hit rates and reduce rebuild times
    3. To make builds run slower but more reliably
    4. To remove all caching and force full rebuilds

12. What does 'hermetic' mean for a build target?

    1. The target uses only standard library functions
    2. The target only runs on Unix-like operating systems
    3. The target compiles without any warnings
    4. The target produces identical outputs given identical inputs every time

13. If a build target depends on the current system time or random numbers, why is marking it cacheable dangerous?

    1. The remote execution will refuse to run it
    2. The cache will fill up too quickly
    3. The build will always fail when cached
    4. Cached results will be inconsistent across runs, creating phantom passes

14. What can AI do when reviewing a Bazel profile?

    1. Install Bazel on your machine
    2. Draft remote-cache configuration diffs with explanations
    3. Execute builds on your CI system
    4. Commit changes directly to your repository

15. Which statement best describes what AI cannot do for build optimization?

    1. AI can analyze BUILD files and identify optimization opportunities
    2. AI can propose configuration changes ranked by expected benefit
    3. AI can decide which specific targets are definitively safe to mark as cacheable without human verification
    4. AI can read build profiles and identify cache miss patterns