Do We Really Need A Test Automation Engineer?
summary
Claims that test automation is an activity not a siloed role, backed by 18 years of experience and repeated automation+isolation equals failure outcomes; early attempts that simply automated a big manual regression suite produced brittle Rube Goldberg scripts and a 12–14 month doomed project that burned money. Best practices shown: pair developers and testers during feature development, target APIs and seams instead of UI-only scripts, put tests into CI with version control and containers, and use the Kling method to decide what, when and who to automate.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:44Test automation journey
- 3:38Failed automation lessons
- 8:04Automation is an activity
- 10:33Manual testers to automation
- 16:35Collaboration over roles
- 19:27Writing tests early
- 24:44Automation in development
- 26:55Tests as conversation
- 29:29Tools are not the solution to automation problems
- 35:18Need a test automation engineer?
related talks
40mPath to Becoming and Developing an Effective Test Automation Engineer
40mThe Test Automation Firehose
47mWhat Successful Test Automation is... and Why it's NOT Testing!
51mRedefining test automation
44mTest Automation as a Development Team
46mOrchestrating Your Testing Process - Coordinating Your Manual and Automated Testing Efforts
13mBeyond the hype: What test automation really costs
21mThe Essential Pillars of Automation Testing for Success