Andreea Cosariu: An AI Experiment: Building a Test Automation Framework Using Generative AI
summary
Built a working end-to-end Java test framework using ChatGPT plus Java 17, Selenide, Gradle and GitHub Actions, with PMD static analysis; iterative prompting fixes address issues like Maven suggestions, circular dependencies and Gradle version incompatibilities, but generated code sometimes is incomplete or loops and needs human fixes. Experiment halves implementation time (roughly 4 days down to ~2), requires coding and CI knowledge, plus isolating problems and giving precise prompts and examples; framework works and is on GitHub and AI requires human supervision.
AI-written summary · about
chapters (10)
- 0:00AI and test automation intro
- 2:37Why start the AI experiment
- 9:02Designing test framework
- 11:11Building with ChatGPT
- 14:23Debugging the framework
- 19:19AI assistance strengths/weaknesses
- 23:56AI vs traditional approach
- 28:08AI amplifies human potential
- 28:50Presenter and contact
- 29:55Q&A
related talks
26mAI and the Future of Testing
01h 26mAutomation of the Automation using GenAI
41mAI-Augmented Testing: How Generative AI, Prompt Eng Turn Testers Into Superheroes
41mAI or Not AI? We Don't (Yet) Believe the AI That Replaces Devs!
01h 02mPANEL - Unleashing the Future: The Impact of Gen AI on Development Productivity
16mCan Test Driven Development be speeded up with Generative AI?
32mHow to Make AI Work for You: A Deep Dive Guide into ChatGPT for IT Professionals
48mTDD & generative AI - a perfect pairing?