"Testing" in Prod Isn't the Only Way: Why Observability Needs Structure
summary
Claims that testing in prod alone is insufficient and that structured observability fixes the gap between code and runtime behavior, by turning observed API traffic into API models that show endpoints, types and diffs. A deployed agent/proxy/CI integration watches test and prod traffic, generates models, annotates pull requests with detected breaking changes (example: international phone number field removed, US format added), and the demo shows automated PR comments and console diffs, demonstrating the approach works end to end.
AI-written summary · about
chapters (12)
- 0:00Intro
- 0:56Overview
- 2:10Jeans background
- 3:22Software development today
- 5:17Software engineering lessons
- 7:19Life is very hard
- 9:38Building and shipping myth
- 16:30Structure observability
- 16:41API models
- 17:29Demo
- 18:20Making a Change
- 19:32Outro
related talks
13mForget Production: Observability a Non-Production Necessity
13mForget Production: Observability a Non-Production Necessity
53mObservability Driven Development
23mA peek into observability from a tester's lens
08m“Observability for testers”
37mObservability-Driven Development
44mObservability: Superpowers for Developers
41mA peek into observability from testers lens