The GPT-ification of Mobile App Testing
summary
Demo shows a no-code platform (sophie.ai) plus OpenAI/Azure LLMs that provisions real physical devices in under 10s (demo uses a Samsung Galaxy S10), records UI flows with ML image recognition, runs a Hilton app scenario end-to-end in under a minute, flags steps over a 1s threshold, detects crashes, and auto-summarizes failures. It also generates runnable Appium code and can synthesize tests from a Confluence spec — SOC2-compliant with in‑house controls rather than wholesale data export.
AI-written summary · about
chapters (11)
- 0:00Introduction to GPT in testing
- 2:15Problems with traditional mobile testing
- 5:29Solving issues with generative AI
- 7:15Demo of no-code testing environment
- 9:03Creating automation with AI assistance
- 11:49Running and analyzing test results
- 13:56Summarizing test outcomes with AI
- 14:37Generating Appium code from tests
- 16:03Generating test cases from specs
- 18:32Future of AI-driven testing
- 19:14Q&A and closing remarks
related talks
18mThe GPT-ification of Mobile App Testing
26mThe GPT-ification of Mobile App Testing
18mExploration of tools, frameworks to take the pain out of mobile app testing!
18mReinventing App Testing with Agentic Apps
35mApp Testing with AI Tools – Make it Easy
11mDragonCrawl: Revolutionizing Mobile Testing with GenAI
28mTesting at the Speed of Thought: QA in the Era of GenAI apps
40mDigitizing Testers