The Feedback Loop from Hell
summary
Slides expose concrete QA anti‑patterns and remedies: TCSD (test coverage stress disorder), false AI trust syndrome, and the “throw everything to QA” bottleneck. Examples include an Air Canada chatbot failure and a cited $1.8 billion brand hit, adoption of DORA and SPACE metrics, integration with Jira/GitHub/Azure via software‑intelligence platforms, and use of artificialanalysis.ai. Recommendations are specific: adopt smarter metrics that actually change behavior, run pair programming on AI outputs, and measure whether fixes cut production incidents.
AI-written summary · about
chapters (17)
- 0:00Vicious circles in orgs
- 1:20Org behavior patterns
- 2:24Vicious circle concept
- 4:12Breaking three-phase pattern
- 5:36Personal pattern-breaking story
- 9:07Feedback loops at work
- 11:15Test Coverage Stress Disorder
- 14:01Metrics changing behavior
- 14:21False AI Trust Syndrome
- 15:48Air Canada chatbot failure
- 16:51AI for quality, not replacement
- 18:38Throw Everything to Same Team disorder
- 20:02Intellectual superpowers vs AI
- 21:25Hiring for quality in AI era
- 22:32Quality enablers with AI
- 24:00Old routines vs AI era
- 24:41Q&A
related talks
22mBeyond Bugs: How AI turns QA into Full-Circle Engineers
41mExits on the (QA) Highway to Hell
24mThe QA spectrum
44mHow AI is Exposing the Existential Crisis of Software Testing
46mQA to QE: Scaling Quality with Ownership, Code, and Culture
43mTesting Agentic AI Applications: Beyond Traditional QA
32mQuality engineering in the AI age: From testing AI to leveraging GenAI
47mPanel Discussion: Who Tests the Machines? QE Leaders on Quality in the Age of AI-Written Code