Richard Seidl
The switch from software development to test automation brings with it clean code and design patterns - but not the instinct to recognize where errors are lurking. If you want to get into automation as a tester without a programming background, you shouldn't start with tools, but with principles: small scripts for recurring tasks, no faith in tools. And while AI-supported testing tools promise miracles, practice shows that when machines write the code, software testing does not become superfluous - but indispensable.
Podcast Episode: From developer to test automation engineer
In this episode, I talk to Benjamin Bischoff about his journey from web developer to test automation engineer - and why this change in perspective has changed everything. Benjamin explains the decisive advantage developers have in test automation, but also where their blind spots lie. We talk about why Selenium has its authorization despite the Playwright hype, how to get into automation as a tester without a programming background and why AI does not replace the role of software testing, but rather dramatically enhances it. In the end, there is a clear thesis: if AI writes the majority of code in the future, testing will become the most critical discipline of all.
"Testing software is also software and must follow exactly the same principles as the application under test." - Benjamin Bischoff
After 15 years as a software developer and trainer, Benjamin Bischoff switched to the field of test automation in 2016. He currently works as a Test Automation Engineer at trivago N.V. in Düsseldorf, where he mainly focuses on backend and frontend testing technologies and pipelines. Benjamin is the author of the book "Writing API testing With Karate" and manages several open source projects in the area of software testing and reporting. He regularly speaks at conferences and writes articles about testing, automation and software craftsmanship on his website softwaretester.blog.
Highlights der Episode
- Test automation code is software - clean code and design patterns are mandatory, not optional.
- Learn principles, not tools: If you know Playwright, you don't automatically understand good automation.
- AI will produce masses of bad software - testing will then become really important.
- The ideal test automation engineer combines development skills with test methodology, not just one of them.
- Selenium is not uncool, it's mature - it defines W3C standards that other tools use.
From developer to tester: a path to better software quality
Many people think that good software is created when talented developers are at work. But quality is more than just well-written code. In the Software Testing podcast, Richie talks to Benjamin Bischoff about how his interest in testing began at school. After training as an IT specialist and working as a web developer, Benjamin ended up in the games industry, where he first came into contact with QA and test processes.
He realized that software not only needs to be built, but also thoroughly tested in order to be error-free and user-friendly. His enthusiasm for tools such as Selenium ultimately led him to switch to test automation. Benjamin has now been working as a test automation engineer for almost ten years, helping to develop quality software.
What is the difference between developers and testers?
Richie asks: What does a developer bring to the table when moving into test automation? Benjamin explains that developers know a lot about clean code principles and design patterns. They make sure that the automation code is just as high-quality as the actual application.
Testers who come from exploratory testing have a good feel for finding errors. But sometimes they lack the in-depth technical knowledge needed to build robust and maintainable automation solutions. The ideal situation is when both sides work closely together: Testers learn from developers, developers from testers. In this way, the team benefits from both skills and can improve software better and faster.
Tools: Will Selenium remain relevant?
Many people are currently raving about tools such as Cypress or Playwright. However, Benjamin sticks with Selenium because he knows the tool inside out and it offers many advantages, especially for browser and mobile testing. He points out that Selenium is not just a tool, but also a standard setter for automation. It continues to develop, even when new tools experience a hype.
It is important not to blindly follow every trend. Those who are deeply involved in the tool know its weaknesses and advantages. And the stability and expandability of Selenium is a clear advantage, especially for long-term test projects.
Learn: Getting started with test automation
How do you get started with testing if you tend to do technical or exploratory testing? Benjamin recommends learning the technical principles, not the tool first. Clean code, design patterns and software craftsmanship are important basics. Small scripts, automation in everyday life, help you get started.
If you simply use the most popular tool, you will learn how to use it, but not the concepts behind it. The programming language is of secondary importance. The decisive factor is an understanding of software development and clean code. If you work as a tester in the software lifecycle, you already have basic knowledge and can deepen it.
AI: Will jobs be replaced?
Artificial intelligence is currently changing many things. Benjamin describes how his team uses various AI tools to support each other: Copilot, Claude or CodeRabbit. However, he considers promises that AI will replace the test team to be exaggerated. Many tools only boil with water. There will be more AI support, but testing is becoming increasingly important as more and more code is generated automatically and quality is difficult to assess.
The switch from developer to test automation engineer shows how important different perspectives are in quality assurance. When both sides work together, better software solutions are created. Technology alone is not enough - principles, collaboration and continuous learning are crucial. The role evolves, but the goal remains: software that works reliably and is a pleasure to use.