Google QA Engineer Interview Questions

In Google, managing multiple services and ensuring quality across the board is vital. A robust tool to aggregate test results will streamline the QA processes.
Problem Statement: You need to create a TestResultAggregator that consolidates test results from various components. Implement methods to add the results and compute the overall results efficiently.
- add_test_result(service_name: str, passed: bool) -> None: Adds a test result for a specific service.
- get_overall_summary() -> Dict[str, int]: Returns a summary of tests, indicating how many passed and failed.
Example 1:
Input: add_test_result('AuthService', True)
Input: add_test_result('PaymentService', False)
Input: get_overall_summary()
Output: { 'passed': 1, 'failed': 1 }
Explanation: One test passed and one test failed.
Constraints:
- `1 <= number of services <= 1000
- Passed results are boolean values.

In various Google Cloud services, understanding response times is crucial for performance monitoring. The ability to swiftly analyze the average response time over a sliding window will aid in identifying and mitigating issues in real-time.
Problem Statement: You are tasked with creating a function that calculates the average response time of Google Cloud APIs over a fixed window size. Implement ResponseTimeMonitor class that maintains a list of response times and computes the average over the last k responses.
- add_response_time(time: float) -> None: Adds a new response time to the monitor.
- get_average(k: int) -> float: Returns the average of the last k response times.
Example 1:
Input: add_response_time(200)
Input: add_response_time(300)
Input: get_average(2)
Output: 250.0
Explanation: The average of the last two response times (200 and 300) is 250.0.
Constraints:
- 1 <= k <= 1000
- `0 <= time <= 10000

As Google scales, it's imperative to effectively manage load testing across services to ensure system reliability. A well-structured orchestrator will help distribute load tests systematically.
Problem Statement: Create a LoadTestOrchestrator class that enables users to schedule load tests on multiple services concurrently. You should define methods for adding services and managing the scheduling of tests.
- add_service(service_name: str) -> None: Add a new service for testing.
- schedule_load_test(test_id: str) -> None: Schedules a load test for all added services.
- get_services() -> List[str]: Retrieves the list of services slated for load testing.
Example 1:
Input: add_service('SearchService')
Input: add_service('AuthService')
Input: schedule_load_test('loadTest1')
Input: get_services()
Output: ['SearchService', 'AuthService']
Explanation: Both services are set for load testing.
Constraints:
- 1 <= number of services <= 100
- Service names are unique strings.

In the world of CI/CD, identifying flaky tests is crucial for maintaining the reliability of Google services. A systematic approach to detect and flag such tests will enhance productivity.
Problem Statement: Design a FlakyTestDetector class to analyze test runs and identify flaky tests based on certain run results. You must define specific methods for logging results and for flagging flaky tests.
- log_test_run(test_id: str, result: bool) -> None: Logs the result of a test run.
- get_flaky_tests() -> List[str]: Returns a list of tests identified as flaky.
Example 1:
Input: log_test_run('test1', True)
Input: log_test_run('test1', False)
Input: get_flaky_tests()
Output: ['test1']
Explanation: The test has inconsistent results, marking it as flaky.
Constraints:
- 1 <= number of test runs <= 10000
- Test identifiers are unique strings.