Beyond the Hype
– Practical Lessons from Using LLMs for Industrial Test Case Generation
Summary:
Large Language Models (LLMs) promise a revolution in software testing, but generating high-quality test cases in industrial settings involves much more than just prompting a model. In this talk, I will share practical lessons learned from an extensive study on LLM-driven test case generation using Google Test. We will explore hidden challenges, critical prerequisites, and subtle engineering decisions that greatly influence compilation success and coverage rates. Attendees will gain a realistic understanding of what it takes to effectively leverage LLMs for industrial-grade testing, moving beyond hype to practical adoption.
Program
-
12:00 — 12:10
Check-in
-
12:10 — 12:45
Presentation and Lunch
-
12:45 — 13:00
Q&A
Speaker
Azeem Ahmad
Ericsson AI Technology Leader Ph.D. and postdoc in Software Engineering from Linköping University
