System reliability often appears to be at odds with developer velocity. The faster your org develops and the more frequently you deploy, the more opportunities you create to cause incidents and outages. But standing still or slowing down are not viable alternatives. In this talk, I’ll suggest some specific ideas, metrics, and UX patterns to bridge this dichotomy. By thinking like a product designer about your testing tooling, you can have happier, more productive developers building more stable, resilient systems.
The focus of this talk will be on high leverage, low engineering effort, suggestions to improve your testing tooling for developers. Some of the tips will be relevant for brand new teams working in small codebases, others will be high-impact for large engineering organizations with their own dedicated test infrastructure or release engineering teams.