How AI is Revolutionizing Quality Engineering: A Practical Look

As someone deeply immersed in the world of software development, I’m always fascinated by how new technologies can streamline different processes in the development lifecycle and elevate every aspect of how we work.

Quality engineering (QE) is no exception. I recently sat down with Stephen Conneely, QE Lead from Fidelity Investments, to discuss exactly how AI is being applied to QE — something I must admit, I didn’t know much about. He had some amazing insight.

One of the biggest shifts in QE over the past decade, Stephen told us, has been the process of “shifting left” — or bringing quality assurance into the earliest stages of development. Stephen explained: “What we mean by that is not breaking the code but ensuring it’s built correctly in the first place. We’re putting in processes to ensure high quality from the very beginning.”

This is a significant shift in how we think about QE — moving away from catching bugs after the fact and instead focusing on embedding quality directly into the DNA of the software as it’s being created.

Stephen explained how AI is also helping QE teams better manage an overwhelming volume of work — a volume that has only been increasing as developers use AI tools like GitHub and PRD-AI to turn out applications faster. There’s just too much data, he said to possibly cover it all with manual testing, and AI is a great way to break that potential bottleneck.

“We’re already seeing gains in automating manual tasks that were previously manual, like test case creation, documentation, and test data creation,” Stephen said. This is allowing teams to focus on higher-level problem-solving and reduce the time to get to the testing stage, rather than getting bogged down in repetitive work.

Another development Stephen highlighted is the potential for predictive analytics to revolutionize QE — what he called the “Holy Grail” of quality engineering. “We can leverage predictive analytics to predict how the code will behave, spot where defects may exist, reuse old data, etc.” he said. “It allows us to step away from maintenance … where we’re not just testing the code but we’re implementing processes for the entire team to follow to ensure quality.”

That kind of proactive approach not only saves time but also enhances the quality of the end product by focusing on the areas that need it most. Consider, for example, how sentiment analysis might help us focus on problem areas that users have flagged, and predictive analytics on error logs and defect reports help us identify high-risk areas before issues even arise.

I was also excited to dig in on another area that Stephen is working on — the concept of self-healing automation. This is where AI really begins to show its full potential, reducing the time QE teams spend on maintenance by 25-30%.

Self-healing code “can adapt to changes in what it’s testing,” said Stephen. “It’s simply a matter of the QE reviewing it saying, ‘Yeah that is expected that that’s not a defect. Let’s change the test case to match.’ And then we click go and that will be that. That job done.”

That means as code changes, automation itself can adapt, reducing the need for constant manual updates. Think about it—automation that essentially takes care of itself. As the code evolves, the test cases automatically adjust to reflect these changes. The time savings is staggering, and it frees up teams to focus on more critical aspects of development.

Even with all these advancements, Stephen was clear that AI cannot fully replace the human element in QE. “AI can’t go and do it by itself. We can’t trust it yet,” he reminded me. This is a sentiment echoed across many industries experimenting with AI: while AI can take on the more monotonous tasks, human expertise is still needed to guide the process and make critical decisions.

Ultimately, the role of the quality engineer isn’t going away—it’s evolving, says Stephen. “The shift is not so much away from technical work but more towards being more personable, and being strong communicators and strong advocates for doing things the right way,” Stephen noted. The key takeaway here is that AI can assist and enhance the work we do, but human oversight and strategic thinking remain crucial to the process.

As we continue to explore the possibilities AI presents for quality engineering, I’m excited by the potential it holds. AI is making our work faster, more efficient, and more precise. But as Stephen emphasized throughout our discussion, it’s essential to strike the right balance between automation and human insight.

The future of QE will be shaped by how we leverage these technologies, but it will always be grounded in the expertise and experience of the engineers who are driving the process forward.

Go from Idea to Code with AI

Schedule a Demo