A testing journey in the age of smart assistants
summary
DellEMC tests shrink a 465 trillion configuration space down to the high‑value set by building Smart Assistants (named cue and Jarvis) that run multiple algorithms against curated defect, test case and sales data; pilots answer 47 prioritized questions, two rolled to production and save about two to three weeks of planning time per platform. Development takes roughly nine months using external data scientists and partners like Capgemini plus heavy data cleaning and provenance work, and initial outcomes meaningfully narrow what not to test.
AI-written summary · about
chapters (17)
- 0:00Intro
- 0:50Agenda
- 1:51Context
- 2:37Test configurations
- 3:59Our journey
- 6:12Analytics continuum
- 7:58Analytics engine
- 9:12Best practices
- 10:01The importance of data
- 11:20What data are we talking about
- 13:58What were our biggest pain points
- 15:21What were our prioritized test cases
- 16:31Finding a data science partner
- 17:49Our cue smart assistant
- 19:18Jarvis smart assistant
- 19:50New UI automation tools
- 20:31Wrap up
related talks
25mWhat’s Our Job When The Machines Do Testing?
44mLeveraging AI and Data Virtualization for QA in ERP Implementations
44mEzanne Grobler, Leela A.Putten: Testing in the Digital Age: A True Story
43mFrom Clickers to Coders: The SDET Evolution Story
43mTesting Agentic AI Applications: Beyond Traditional QA
06mTriple Threat: AI Technologies Reduce Testing Costs
01h 04mImpact of AI on #SoftwareTesting: Are Testers Ready? 🔮
14mPrompt Engineering Impacts to Software Testers : Staying in Demand