Topic
Engineering Practice
Page 2 of 30 articles about Engineering Practice, newest first.
Browse the complete article archive →Drain a Queue Backlog Without Overloading the Services Behind It
Calculate backlog growth and conditional drain time with a filled recovery example, aggregate downstream call limits and an explicit retreat decision record.
Is This a Breaking Change? Apply Semantic Versioning to a Real Release
Classify ten concrete API changes, inspect a real local compatibility mismatch, and record a version decision with clear test and release boundaries.
Make a Pull Request Small Enough to Review Without Hiding Its Context
Make pull requests easier to review with a focused split, clear dependencies and scoped tests. Use a complete sequence and two filled review descriptions.
Measure Developer Productivity Without Ranking People by Commit Counts
Use a completed team measurement agreement and synthetic two-window review to compare commit activity, review waits, quality, and reported focus fairly.
Plan a Load Test Around Real Traffic Patterns and a Stopping Rule
Plan a load test with explicit arrivals, request mix and a stopping rule. Inspect an actual bounded local run, its rejected requests and capacity limits.
Rebase or Merge Before a Pull Request? Preserve the History Your Team Needs
Choose rebase or merge before a pull request using shared-history checks, actual commit graphs and a tested conflict resolution that preserves both changes.
Remove a Temporary Feature Flag After the Rollout Is Finished
Remove a temporary feature flag with a filled lifecycle record, checked before-and-after code example and explicit environment, fallback and recovery gates.
Review Third-Party SDK Data Collection Before Adding It to an App
Review SDK data collection, permissions, configuration, and unresolved evidence with a complete Android example that separates vendor claims from actual tests.
Roll Out a Feature Flag with Targeting, Metrics, and a Stop Condition
Roll out a feature flag with stable targeting, clear metrics and named stop authority. Review a complete fictional stage packet and a failed-guardrail record.
Run a Release Readiness Review with Evidence for a Go or No-Go Decision
Run a release readiness review with a completed evidence packet, no-go decision, remediation review, and approval limits that stay separate from deployment.
Set a Cache Lifetime Around How Stale the Data Is Allowed to Be
Set public and private cache policies from acceptable update lag, then inspect an executed local timeline across expiry, revalidation, and origin failure.
Test with Realistic Data Without Copying Customer Records into Development
Design realistic synthetic test data with a constraint contract, reproducible boundary records and checked filtering cases, without copying customer rows.











