BearingPoint’s Karl Byrne, Holly Daly and Fiona Eguare discuss the implications of AI for applications engineering and how it has particularly affected graduates.
The widespread integration of advanced AI technologies into technology workplaces around the world has changed the working life of many, but especially in software teams.
“Over the past few years, software engineering has undergone some of the most significant changes I’ve seen in my career,” said Karl Byrne, director and head of software development at BearingPoint Ireland.
“As the industry navigates traditional cloud and DevSecOps transformations, the advent of generative AI represents a fundamental shift in the way we think about, build and secure software.”
Byrne tells SiliconRepublic.com that what strikes him most is how widespread the change is. “It’s not just limited to experts or one team – it affects every part of how we deliver software.”
However, he added that the fundamentals of the area have not changed, stressing that a strong understanding of technology, sound design principles, and a focus on safety and quality are “as important as ever”.
“If anything, AI has raised the bar, because engineers now need to carefully evaluate work done with AI above everything else they do,” he explains.
For graduates, Byrne says, the introduction of AI into the role has encouraged a “complete revolution” of everyday roles.
Responsible use
Holly Daly, technology analyst at BearingPoint Ireland, says the growing use of AI highlights the importance of using these tools carefully and responsibly – especially for graduates and budding software engineers.
“While AI can greatly improve productivity, graduates should avoid over-reliance on it and continue to build on the basic skills they have developed,” he said. “AI should be used as a supporting tool to improve efficiency and quality rather than as a substitute for your technical understanding and critical thinking.”
He explains that it is very important for a graduate to demonstrate that they understand the solutions they are delivering and are not just relying on AI.
“From my experience as an undergraduate working on an AI-driven project, I have had the opportunity to work with several AI tools, testing and recommending them,” he said. “At the same time, I am focused on learning to improve my skills so that I am not dependent on AI. This approach has allowed me to benefit from AI, while allowing me to work confidently on my own.”
Daly says BearingPoint’s graduate program adapted to AI-assisted engineering by exposing graduates to AI early on and integrating it into both their training and project experience.
“During the onboarding, graduate students are given exposure to AI through dedicated lectures and discussion sessions, including AI walkthroughs that highlight its strengths, limitations, and potential use cases. These sessions help build an initial understanding of how AI can support technical and non-technical tasks, while emphasizing the importance of appropriate use.”
Fiona Eguare, also a technology analyst at BearingPoint Ireland, says the process of incorporating AI technology into the engineering team has many steps – starting with research and testing.
“We evaluated the available tools and tested the ones that seemed to be the best fit for our needs. This allowed us to compare them, make sure they fit our use cases, and evaluate the advantages they offer over traditional tools and methods,” he said.
“Once the most useful tools are identified, we share our findings across the team and the wider company, and integrate the tools into the project where appropriate.”
Eguare says that while everyone involved has been enthusiastic and open to incorporating AI throughout the software development lifecycle, it’s very much an “ongoing effort.”
“As the tools continue to improve, it will be important that we continue to improve capabilities and monitor their security, to ensure they are always ready for us.”
AI-driven changes
Both Daly and Eguare say the incorporation of AI tools into their work life has had some benefits.
“One of the clearest results for me,” said Eguare, “is the increased ability of developers. With the help of AI productivity tools, some of the most tedious and time-consuming development tasks can be completed much faster.
“These tools can be very helpful when debugging. Although they can sometimes miss the mark on this, some productivity AI tools do a great job of understanding the context of the project and the codebase, making them great at finding the source of bugs.”
Daly found that tasks such as writing new code, refactoring existing bugs and fixing bugs have become “much faster and more efficient” with the support of AI tools.
Along with the benefits, both also recognize the potential pitfalls of the technology.
Eguare highlights the cybersecurity risks of technology, saying it has made it easier for attackers to exploit vulnerabilities, while Daly says AI has changed the role’s requirements.
“The role is no longer just about writing code, but also about reviewing, validating, and improving the work generated by AI,” Daly said. “Software engineers need usability and analysis when evaluating whether proposed AI code is correct, secure, maintainable, and appropriate for the problem being solved. As a result, strong technical understanding and critical thinking are more important than ever.
“Overall, while AI can be an effective productivity enhancer, it’s important that software developers don’t let it take over, as the onus is on them to ensure that the final solution meets the required standards.”
Human oversight
What is always important in using productive AI tools in software engineering, according to Eguare, is human supervision.
“When we work as a team on large and important projects, oversight is important; its importance cannot be overemphasized,” he said.
“Lack of monitoring can lead to problems, such as bloated code or serious vulnerabilities that spill over into production.”
Eguare explains that in order to deal with these problems, it is important to use “high-quality information, which specifies expectations in terms of quality and safety”, and testing.
“Along with traditional testing, tools that directly address common problems with AI-generated code can be very helpful here,” he said. “We also rely on automated quality CI/CD pipelines and security scanners to enforce consistent standards and catch problems early – especially important when AI accelerates code changes.”
Another issue he highlights is that when a lot of programming is done without human supervision, it can be “really hard” for a developer to debug or understand the codebase.
“While AI can also help with this, staying familiar with the architecture of the program can help ensure that the code remains clean, secure, and up to date as changes are made.”
Don’t miss out on the information you need to succeed. Sign up for Daily BriefSilicon Republic’s digest of must-know sci-tech news.
