You are currently viewing a placeholder content from Podigee. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More Information
This podcast episode is about test automation and the use of machine learning to find defects faster and integrate tools. Unsupervised Learning and Test-Based Modeling enable more efficient testing processes. Augmented Testing combines human knowledge and automation for more accurate results. The future of test management is discussed in terms of the role of AI in test automation and further optimization.
"I come in in the morning and instead of finding 200 failed tests, I find 7, maybe 8 causes" - Thomas Steirer
Thomas is a test automation architect from Vienna, and brings over 15 years of experience in this field. He has developed numerous automation frameworks and solutions across a wide range of industries and technologies. His focus is on building scalable and sustainable solutions that are primarily designed to deliver valuable information. In his work at Nagarro, he assists customers in implementing and optimizing test automation, teaches at universities in Austria and is co-author of the book "Basiswissen Testautomatisierung".
Highlights in this episode:
- Thomas and I have known each other since our first test projects in Vienna
- Thomas is CTO at Nagaro and an expert in artificial intelligence and test automation
- We talk about a research project investigating how machine learning and AI can be used to improve test automation structures
- The aim is to draw new knowledge from existing information and identify the causes of errors
- We discuss how to write new test cases and use visualizations to increase test quality
- Thomas talks about a research project that involves visualizing test portfolios and gaining valuable insights from them
- The importance of making life easier for testers by minimizing repetitive tasks is emphasized
- Thomas talks about the future of testing and how machine learning can help increase test depth and coverage
- We discuss the role of testers in a test landscape supported by AI and come to the conclusion that human testers are still indispensable.
Contact Thomas:
How artificial intelligence is redefining test automation
TLDR: The AI revolution in test automation is fundamentally changing the way we test software. By using AI and machine learning, we can optimize and automate testing processes, resulting in more efficient and effective testing. These technologies not only enable us to identify and fix errors faster, but also to generate new test scenarios and expand test coverage.
The convergence of AI and test automation
In recent years, the landscape of software development has changed dramatically. A driving force behind this transformation is the integration of artificial intelligence (AI) into test automation. As I discussed in this episode with Thomas Steirer, CTO at Nagarro, we are at the beginning of a revolution that has the potential to fundamentally change the way we approach software testing. The integration of AI into test processes not only makes it possible to optimize existing test scenarios, but also to open up new ways of identifying and eliminating errors.
A beer garden conversation
The idea for the research project arose from a seemingly banal moment - a conversation between colleagues in a beer garden. But as is so often the case, it is precisely these unexpected moments that lay the foundation for innovation. Thomas described this moment to me as a point of self-reflection and creative exchange, which ultimately led to the decision to use machine learning (ML) to optimize test automation processes. This step marked the beginning of our journey together into the world of AI-supported test automation.
From theory to practice: initial successes
Our research project focused on applying machine learning methods to existing test automation structures. A key objective was to extract new knowledge from existing data and identify the causes of errors more efficiently. Thomas explained to me how, by analyzing log files and using algorithms, we were not only able to identify the causes of errors more quickly, but also generate initial suggestions for new test cases. The success of this approach confirmed our conviction: AI can add significant value to test automation.
Visual analysis and model-based testing
Another breakthrough in our project was the development of visual analysis tools and model-based testing approaches. By merging all technical test steps in a graph, we were able to create a visual model of the application structure. This model not only enabled us to quickly identify redundancies and error hotspots, but also to derive new test scenarios - a milestone in our research into the integration of AI into test automation.
in future Automated generation of test cases
Looking back at what has been achieved and ahead to what is still possible, it seems clear that the future of test automation lies in the further integration of artificial intelligence. Thomas outlined exciting prospects such as the automated generation of test cases based on AI-generated assumptions or even explorative tests controlled by AI. These developments promise an unprecedented increase in efficiency in the testing process.
Man vs. machine in testing
Finally, of course, there was the question of the role of humans in the age of AI-supported test automation. Despite all the progress, Thomas and I agreed that the human factor remains irreplaceable. It's not just technical know-how, but also an understanding of context and nuance that makes for high-quality testing. The integration of AI offers enormous support possibilities and can make our work easier - but in the end, it is still the human factor that determines quality.