Replay Production Traffic Locally: Instantly Spot & Fix Breaking Changes
summary
Open-source tool replays production traffic locally by adding a single SDK annotation and an IDE plugin, capturing per-line timing and downstream calls, saving replays as zipped resources and running them via JUnit/Maven. Demonstrated workflow shows how recorded replay production traffic locally sessions become repeatable integration tests, how an AI-generated runtime PRs feature edits code and runs the replays to validate fixes, and that the project already has 10,000+ downloads and aims to cut the 640x higher cost of fixing production bugs.
AI-written summary · about
chapters (10)
- 0:00Introduction to replaying production
- 2:49Why bugs reach production
- 7:43Production vs local differences
- 10:51Introducing the tool
- 13:42Live view and method execution
- 18:31Recording and replaying traffic
- 22:27AI code fixes with replay data
- 29:19Architecture and performance
- 34:59Extending to other languages
- 39:44Q&A and closing
related talks
46mRecord-Replay API Test Cases & Data Mocks without writing code
50mAutomate Finding & Fixing Bugs with Replay QA
17mBe Kind Rewind with Test Replay
29mFrom Real Traffic to Reliable Integration Tests with AI
14mYour Agent Failed in Prod. Good Luck Reproducing It.
33mContainerized Time Travel: Replicating Production Performance
29mBlink and you’ll miss it: Stories from discovering most annoying bugs and test flakes
36mMaximizing Debugging Efficiency in CI CD Pipeline with Cypress Test Replay