Test Intelligence: Finding More Bugs in Less Time with AI
summary
Five concrete, reproducible analyses are shown: historic bug heatmaps that locate untested hotspots (one product had front end gaps that caused most production fixes and required foundation work before the next sprint), test gap analysis that flags unexecuted changed code (a 30‑person six‑month release shipped untested areas and was delayed three weeks), coverage‑based test selection and impact analysis that reduce runtime dramatically — 6% of execution time finds ~80% of bugs — and ML attempts at defect prediction which so far fail at method level.
AI-written summary · about
chapters (6)
- 0:00Test intelligence challenges
- 11:39Analyzing bug hotspots
- 16:00Untested changes in production
- 25:20Smart test selection
- 37:26Predicting bugs with AI
- 42:39Summary and applications
related talks
50mTest Intelligence for Architects
42mBeyond “Just Try Harder”- Three effective strategies to tackle bugs
40mFrom Bugs to Brilliance: Leveraging AI in Modern Software Testing
06mTriple Threat: AI Technologies Reduce Testing Costs
41mHow AI is transforming software testing
10mBeyond “try harder”: Effective strategies to tackle bugs
20mWill More Test Code Really Reduce Bugs…?
26mThe Impact of Testing on a DevOps Pipeline